At the top of their conclusions, in a 2009 paper in the Journal of Personality and Social Psychology, two researchers set the familiar phrase aside. “Although it is said that travel broadens the mind, in the current studies, we found a robust relationship between living in and adapting to foreign countries and creativity.”
Read it twice. They had gone looking for the old saying and had come back with something next to it, but not it.
Two columns instead of one
William Maddux, then at INSEAD, and Adam Galinsky, then at Northwestern, ran five studies. The methodological move that makes them useful is small, and almost nobody makes it. When they asked participants about time in foreign countries, they asked twice: how long have you lived abroad, and how long have you traveled abroad. Two columns, not one.
The Artful Age
A weekly letter on aging well, family across generations, and the creative life after the kids leave home.
In the first study, 205 MBA students were emailed a link to the Duncker candle problem. You see a candle, a pack of matches and a box of tacks, and you have to fix the candle to a wall so it burns without dripping wax. The answer is to empty the box, tack the empty box to the wall, and stand the candle in it. The problem tests whether you can stop seeing a box as a container for tacks.
Just over half of them got it. Controlling for age, gender and nationality, the years someone had spent living abroad predicted solving it. The weeks they had spent traveling abroad did not. The travel coefficient actually came out slightly negative and just crossed the significance threshold, a result the authors report once and never mention again, and which is too small and too strange to build anything on.
The second study swapped the puzzle for a person. One hundred and eight MBA students negotiated the sale of a gas station in pairs, in a scenario rigged so that no price can work: the most the buyer may pay is less than the least the seller may accept. The only deal available requires inventing a second issue, future employment for the seller, that neither side’s briefing suggests. Just over half the pairs found it.
This time the controls included all five major personality traits and final class grades. Openness to experience predicted the creative deal, as you would expect. So did extraversion. Agreeableness predicted it negatively. On top of all of that, time lived abroad still predicted it, while time traveled abroad produced a coefficient of almost exactly zero.
It was never the years
The fourth study is where the claim changes shape.
One hundred and thirty-three MBA students at a European business school, between them holding forty nationalities, reported how long they had lived in each foreign country and, for each one, how much they had adapted themselves to the local culture. Then they were given the candle problem.
Time abroad predicted solving it, controlling for the five personality traits, age, gender, grades, admission test scores, languages spoken fluently and number of countries lived in. But when the adaptation score was added to the model, adaptation predicted creativity and time abroad was no longer significant. The duration had been standing in for something else all along.
Two experiments went at the same idea from the other side. In one, 65 undergraduates at a French university, all of whom had already lived abroad, were randomly assigned to spend a few minutes writing about living abroad, traveling abroad, a day in their hometown, or their last trip to the supermarket, and then took a word-association creativity test. The living-abroad group scored higher than each of the other three. The traveling-abroad group did not differ from the hometown group or the control group.
In the other, 102 undergraduates at a French university, again all of whom had lived abroad, wrote about adapting to a foreign culture, observing a foreign culture, learning a new sport, or nothing, and were then asked to imagine a planet very different from Earth and draw a creature they met there. Three raters who knew neither the hypothesis nor the conditions scored the drawings. The people who had written about adapting drew creatures judged more creative, less like Earth animals, and carrying more unusual sensory organs. The people who had written about observing a foreign culture scored no differently from the people who had written about taking up a sport, or about nothing at all.
Observing and adapting are not the same act, and in that study only one of them moved anything.
How much weight this can hold
Less than the story wants it to. The authors say so themselves: “the current data do not demonstrate that living abroad causes permanent changes in trait-based creativity.” Three of the five studies are correlational. The other two prime a temporary state in people who had already lived abroad, which tells you nothing at all about anyone who has not.
The obvious alternative explanation is that creative people move abroad rather than the other way round. Maddux and Galinsky’s answer is decent without being decisive: creative people would presumably travel abroad more as well, and travel never once came out positive.
Two further cautions belong here that the authors do not raise. The participants are elite MBA students and French undergraduates, which is not the world. And those two experiments are small priming studies from 2009, with roughly sixteen and twenty-five people per condition, the design and the vintage that has fared worst in psychology’s replication reckoning. No independent replication or meta-analysis of this specific effect appears to exist.
Meanwhile, something smaller and much better measured
Eleven years later, in a different field entirely, Aaron Heller, Catherine Hartley and colleagues asked a narrower question and answered it with better data.
They recruited 132 young adults in New York and Miami and tracked their phone locations continuously for three to four months, a mean of 104 days. Every other day, at a random moment, a mood questionnaire arrived by text message or app. After exclusions for too few responses, 122 people were analyzed.
For each day they computed roaming entropy, a measure lifted from research on rodents exploring enclosures. It is high when you visit many distinct places and spread your hours evenly among them, and lowest if you spend the whole day in one spot.
Positive mood was higher on high-roaming days. Negative mood had no relationship with it at all, which is worth noting: this is not a study about feeling less bad.
The useful part is the control variables. The model also contained the distance traveled that day, the city, the day of the week, the hour the survey was answered, and the temperature and rainfall where the person had spent most of their time. Of that entire list, only day of the week mattered. Distance did not. It was not how far people went that tracked with their mood. It was how many different places.
Pushing further, the researchers counted, for each day, the locations a person visited that they had not been to before during the tracking window. That count predicted mood more strongly than roaming entropy did, and when both were entered together, novelty survived and roaming entropy did not. The same pattern held for a measure of how demographically varied the visited places were.
There was also a loop running both ways. The previous day’s novel locations predicted today’s mood, and today’s mood predicted tomorrow’s novel locations. The authors call it an upward spiral, then say plainly that establishing cause would require a study that manipulates one side of it, which theirs did not.
Their other limits are worth keeping in view. Everyone was between 18 and 31 and living in a large American city. The tracking window was only long enough to say a place was new to that person during the study, not new in their life, and the authors ask for a design that can tell the difference. They also flag an open question that matters for anyone whose movements are not their own choice: whether the effect depends on the roaming being voluntary.
Two different goods
These are unrelated teams in unrelated fields, measuring unrelated outcomes, and neither paper cites the other. It would be sloppy to fuse them into a single claim. Laid side by side, though, they cut the question in a way that is hard to unsee.
One thing going somewhere new gives you is novelty, and novelty appears to lift mood on the day. That effect runs on the number of unfamiliar places rather than the miles, which means it is genuinely available on a Saturday afternoon, in a part of your own city you have never walked through.
The other thing, the one that the phrase about broadening the mind is groping for, is harder to come by. The evidence for it is thinner, mostly correlational, and drawn from narrow samples. But such as it is, it does not attach to visiting. It attaches to adapting, and adapting requires staying somewhere long enough that not adapting stops being an option.
What to do with that
Almost nothing, practically, and pretending otherwise would be dishonest. Most people cannot relocate to another country, and nothing in this research says that those who have not are less creative or less broad-minded. Maddux and Galinsky point out themselves that the domestic version of the question, moving between genuinely different regions inside one country, has simply never been tested.
What it does offer is a more accurate expectation. A trip is not a smaller dose of an expatriate life, any more than listening to a language is a smaller dose of speaking one. The holiday delivers its own good, and the phone data describes that good precisely. The larger change seems to be bought with the part of a foreign stay that nobody photographs: the stretch where the rules are not yours, you keep getting them wrong, and you have to change in order to stop getting them wrong.