Would you rather save 90 percent with a product discount, or 91.27 percent? New research suggests that our minds are wired to prefer nice, round numbers over irregular ones, even when the irregular option means a better deal overall.

This taste for numerical aesthetics was revealed in a study involving 1,552 participants across six different experiments. The volunteers were tested on their responses to, and assessments of, rounded and non-rounded numbers when put up against each other.

Whether it's a marketing splash on a billboard, product information on packaging, or a public health campaign being run by a government, these findings can be applied in all kinds of ways to avoid confusion and to nudge people in the right direction.

"Numbers have a language and give non-numerical perceptions," says behavioural economist Gaurav Jain, from the Rensselaer Polytechnic Institute in New York.

"When we use specific numbers, the evaluations decrease. There was no apparent reason for this kind of behaviour, and this was incredibly surprising."

Seeing unusual numbers is jarring for us, the researchers think, potentially requiring more brain power to process. What's more, non-rounded numbers are more likely to be compared to ideal figures – like 100 percent – just to make better sense of them.

In our attempts to understand irregular numbers that don't end in a zero, we give them negative connotations, according to the study – they are unable to match up to the ideal rounded figures that we instinctively compare them against.

All this should be taken into consideration when using and framing numbers, the team behind the study says – especially if people need to take action as a result, which is the case when it comes to health advice given out during the current coronavirus pandemic.

"The extensive use of attribute framing in marketing, organisational behaviour, and public policy communication and the robustness of the effects in experimental settings make it one of the most important and frequently studied phenomena in the field," says Jain.

"Managers and public health officials should be careful when using non-round numbers, because the use of this approach in communication messages may decrease the subjective evaluations of the target on the associated attributes."

It's a fascinating topic and one that is more nuanced than you might think. Other studies have shown how non-rounded numbers can make those figures seem more trustworthy and reduce the inclination to negotiate, for example.

While plenty of previous studies have looked at the way numbers are described and the words used alongside them, the new research tackles the numbers themselves – an area that hasn't been thoroughly examined before now.

And that means there's much more to come in this field of study: in this latest research, test participants were only assessed using standard economic research questions rather than with specific scenarios, which is one option to explore in the future.

"Our studies lend support and offer an elaborated process account for the attention-association-based reasoning for framing effects in general, which adds to the scarce literature on processes underlying framing effects," says Jain.

The research has been published in Organizational Behavior and Human Decision Processes.