When I joined a group of children at my primary school to tempt a classmate to sample gleaming red berries from a shrub in the playground, we were taking a risk: the berries might be delicious and nutritious - which is what we hoped - or they could be poisonous and make her very sick. Nobody really thought about that.
It was a gamble, but not one we were taking with our own health. Instead we were persuading another child, who was less able to calculate the risks, to take the wager for us. In our defence, we were only 6 - but it was a brutal Lord of the Flies-style experiment.
Gambling, we're told, is nearly always bad. And schoolchildren should be warned about its perils. You could lose your health (if you gamble with drugs and drink); your money (because you become addicted and don't know when or how to stop); your home (because you spent all your money); and your job (as the above has rendered you homeless, penniless and a physical wreck).
Statistics from GamCare, a charity that supports gambling addicts, showing the increase in the numbers of young people with gambling problems make terrifying reading. But it has now been discovered by a team of neuroscientists at the University of Bristol that gambling may have its uses in the classroom, by incorporating the notions of risk and uncertain outcomes into lessons.
Paul Howard-Jones, who led the research, says this is proving more effective than the traditional carrot-and-stick approach, which works on the simple structure of "get the right answer, get a reward". Humans are wired in a way that makes us more motivated to learn when we are offered even the chance of getting a reward, he explains. So perhaps planting the perception in a child's mind that they are being allowed to take a risk is a clever piece of psychology. They think they have everything to play for.
Children are by nature competitive. They are often less sensitive to the feelings or well-being of others than most adults, so such systems need careful management. A teacher, for example, would have intervened all those years ago in our playground experiment. In the end our friend was OK, despite the fact that the berries were not edible.
Jo Knowsley is acting editor of TESpro