I Trust You to Do This

For decades, Professor Paul Harris has been studying trust, especially how young children make decisions about whom to trust. But now, in a new paper published in the journal Nature Human Behaviour, Harris and his colleagues have flipped the conversation. “What happens when we don’t invite the child to trust,” he says, “but we invest trust in the child?” More specifically, they wanted to know if young children would cheat less if an adult showed trust in them.

What Harris found, along with researchers at Hangzhou Normal University in China and the University of Toronto in Canada, is that yes, simply asking young children to help with even a small task and praising them for being a trustworthy helper, and then conveying that they would trust them in the future led to less cheating by 5- and 6-year-olds when compared to children who were not given the same “trust messages.” 

Says co-researcher Li Zhao, a professor at Hangzhou University, “It seems that even at a young age, children understand the value of trust and are willing to behave more honestly in response to feeling trusted by others.” 

The team figured this out after conducting five experiments with more than 300 kindergarteners in China. In the first experiment, each child was walked down the hall by a teacher to another room to take a short math test. On the way, the teacher asked the student to help by holding an important envelope that had the answers to the test. The student was praised for their help and reminded that they were trusted not to cheat. The teacher also said they would continue to trust the student in the future. The four subsequent experiments were variations on this test, leaving out one or more elements — the teacher didn’t praise the student or didn’t say they were trusted not to cheat, for example. In the fifth experiment, the person administering the test also wasn’t the teacher. 

At some point, before the student finished the short test, the teacher stepped out of the room briefly, leaving the answer key on a table near the student, easily within view. A hidden video camera revealed whether the student peeked at the answer key, especially for the last question, which was particularly difficult. When students received the maximum level of trust messages, the cheating rate was 34%. At the other end, when students were told they were trusted not to cheat but were not told they were trusted to be helpers, more than half cheated. Nearly half cheated when one teacher asked for help, and another gave them the test. 

Harris speculates that the results they found — that young children who were directly asked to help and received trust messages were less likely to look at the answers when the teacher left the room — are partly evolutionary.  

"As a social species, establishing mutual trust would have conferred survival advantages for our distant ancestors,” he says. “We could speculate and say that human beings have learned that it’s good for you to respond in a trustworthy fashion if somebody has trusted you just in the same way as you trust somebody who has been trustworthy.”

Zhao says that for parents and educators, realizing the power that trust, rather than threats or punishment, can have in nurturing integrity in children is valuable.

“Our results demonstrate that clearly articulating trust can reduce dishonesty,” she says, “and this highlights the need for parents and teachers to consciously show and express trust to children through everyday asks. Rather than assuming children infer trust through actions alone, directly conveying it may be key. We recommend teachers intentionally entrust students with tasks and responsibilities, no matter how small, while explicitly telling them. ‘I trust you to do this.’”

Harris says there’s also a broader general lesson that can be learned. 

"If you treat a child in ways that they understand to be positive, you might well be helping them to put their best foot forward,” he says. “Not just with respect to cheating, but perhaps in other domains.” He says that thinking about the results of this study reminded him of an experience back at Oxford University, where he studied and taught before coming to Harvard.

“I used to teach in Oxford, which is divided into colleges. And each college is, up to a point, fairly autonomous in its mini democracy,” he says. “I noticed that from time to time, there would be murmurs of colleagues who were a complete pain in the neck, and sometimes, the president would adopt the strategy of giving this person some responsible job. Oftentimes it worked. This person, being endowed with responsibility, stopped being such a pain in the neck and stepped up to the plate.”