A new study finds that when asking someone to swear honesty, phrasing matters.
27 January 2025
By Emma Young
Could simply making a promise to be honest encourage people to behave in an honest way, even when there's a temptation to lie? A recent study in Nature Human Behaviour suggests that it can — but the precise nature of the promise matters.
Janis H Zickfield at Aarhus University, Denmark, and colleagues used a tax evasion game to measure dishonesty among a total of more than 21,000 participants from the UK and US. These participants completed online tasks to earn money, then were asked to report how much they had made. These earnings would be taxed at 35%. To avoid this high toll, however, the participants could lie about their earnings, and so reduce the amount of tax that they had to pay, without any repercussions.
At some point during the game, they were also all asked to make one of a total of 21 'honesty oaths' — or, for the control group, to make no oath at all. For 20 of these interventions, the most basic oath (eg. 'I hereby declare I will provide honest information') followed another more complex statement which emphasised anything from the need to earn the trust of fellow citizens to the idea that honesty is a community responsibility.
The timing of these oaths also varied. Some participants made them before embarking on their tasks, while others did it just before declaring their income. The format varied, too, with some participants being asked just to tick a box to indicate their endorsement, while others had to type the statement in full.
When the team analysed the results, they found that overall, a quarter of participants under-reported their income to some degree. Almost a third of this group falsely declared having received no income at all.
According to further analyses, not all honesty oaths are created equal. Of the 21 oaths, 11 had no impact on income honesty. These were generally of a type that referenced the individual's moral character, or alluded vaguely to their social responsibilities — such as 'I am an honest person' or 'Honesty fosters trust in society'. The list of failed oaths also included statements that emphasised the threat to a person's self-image or the guilt they would feel if they were dishonest.
However, the remaining 10 oaths did boost income honesty. The most effective of these almost halved the lost tax from to 11.6%, compared to 21% in the no-oath control group. The winning oath explicitly linked honesty to reporting income — 'I hereby declare that I will provide honest income when reporting my final income from the sorting task.'
Other effective oaths included statements that framed honesty as being all or nothing ('Either the reporting is honest or it is not') or specified social costs of dishonesty, such as a reduction in funds that would go to the Red Cross. These clear, specific statements made it harder to the participants to justify lying about their income to themselves, the team thinks.
These results fit with work in motivational psychology, finding that reducing the ambiguity of a goal helps people to achieve it, and with findings that specific rules are more effective at reducing dishonesty than general rules.
The team also found that the method of making the oath had some impact. Though this did not hold across the board, when the most effective oath plus one other, on responsibility ('I understand it is my responsibility to report honestly') were typed out, rather than endorsed by ticking a box, they were more effective at encouraging honesty. Presumably this is because these participants were forced to think more about what they were endorsing.
The timing of the oath didn't make much of a difference to honesty, but oaths were slightly more effective when they were made immediately before a participant reported their income. "This suggests that connecting the honesty oath as closely as possible to the behaviour to be influenced might be helpful," the team writes.
It's worth noting that none of the oaths completely eliminated dishonesty. In the real world, though, just reducing it could have major implications. In the UK, tax evasion costs the government an estimated £5 billion per year, for example — though as anyone in the UK who has completed a self-assessment tax form will know, one of the final steps is to tick a box stating that the information given in the form is honest. Concrete and specific honesty oaths, required at the time a decision is made, might be effective in other areas of life, too, however — in workplaces, for example.
"While other honesty interventions, such as audits or punishment, might be effective in specific contexts, the current study offers evidence that honesty oaths can serve as low-key, cost-effective interventions to curb dishonesty," the researchers conclude.
Read the paper in full:
Zickfeld, J. H., Ścigała, K. A., Elbæk, C. T., Michael, J., Tønnesen, M. H., Levy, G., Ayal, S., Thielmann, I., Nockur, L., Peer, E., Capraro, V., Barkan, R., Bø, S., Bahník, Š., Nosenzo, D., Hertwig, R., Mazar, N., Weiss, A., Koessler, A.-K., & Montal-Rosenberg, R. (2024). Effectiveness of ex ante honesty oaths in reducing dishonesty depends on content. Nature Human Behaviour. https://doi.org/10.1038/s41562-024-02009-0
SOURCE:
