What is the best way to estimate risks or make important decisions at work or at home? A new paper from researchers in Nottingham and Amsterdam has concluded that it is usually better to adopt the old adage, 'sleep on it'.

"If we take a break for a few nights this allows us to gain a fresh perspective," explains Dr Dennie van Dolder, who was a postdoctoral researcher in the ESRC Network for Integrated Behavioural Science (NIBS) project, University of Nottingham.

The new method draws on the 'Wisdom of Crowds' principle, a well-established decision-making technique that has been used to improve the accuracy of economic forecasts, medical treatments and weather forecasting.

"Many human decisions, whether in business, politics, medical or personal domain, require the decision-maker to estimate unknown quantities," says Dennie van Dolder. "According to the Wisdom of Crowds principle, more accurate estimates can be obtained by combining the judgments of diverse collections of individuals than by experts alone." The principle is based on the statistical fact that the average of a number of imperfect estimates is more accurate because it reduces the effect of errors.

The new research published in Nature Human Behaviour explores whether the same principle applies to single individuals faced with making a decision.

The so-called 'Wisdom of the Inner Crowd' could be applied to personal decisions such as choosing a university course or whether to move house, or to decisions requiring extreme expertise such as the best way ahead in a new business project where input from the uninitiated would not be feasible.

The study findings are based on the results of a series of promotional competitions organised by a state-owned Dutch casino chain over seven-week periods at the end of 2013, 2014 and 2015. The participants had to guess how many small objects, such as pearls and diamonds or casino chips were in a plastic container shaped like a giant champagne glass. The prize of €100,000 each year was shared by the 16 winners in the first year and awarded to single winners in the second and third years. All the winners guessed the correct number, which varied from 12,564 to 23,363.

The guesses were registered on terminals via a voucher code when they arrived at the casino. Studying the pseudonomised data from the guesses they made on each visit, the researchers found that averaging multiple judgments from a single person does indeed improve accuracy, especially if there is a gap of about two weeks between consecutive guesses. "We think this is because the time lapse allows people to forget their first attempt and to look at a problem with a fresh perspective," Dr van Dolder explains.

However, the findings also suggest that the greater accuracy from the average of multiple guesses from a single individual is considerably lower than the gain from guesses from two different people. This has important implications for decision-making: it is better if two people work on the same two projects rather than specialising on independent projects. However, if that isn’t possible, it is a good idea to sleep on it rather than make impulse decisions.