Like paying taxes, or rain on the weekend, computer games are now a fact of everyday life. From the humble beginnings of Space Invaders and Pac-Man, games now sport mind-boggling 3-D graphics and more villains to shoot than there are fish in the sea.

Most people think children get about as much educational benefit from playing these games as they do from watching cartoons, but Richard Noss, for one, does not believe this is true.

A professor of mathematics education at the University of London’s Institute of Education, he says that children come into contact with plenty of very powerful ideas when playing games - but how these work are hidden in the workings of the program. If only children had a way to get inside the games, and think about how they work, a lot of learning could be unlocked.

Noss is co-director of Playground, a three-year pound;750,000 European Union research project involving commercial and academic members who aim to develop a computer environment where children can create their own games. These include software company Logotron, the Institute of Education and universities in Sweden, the Slovak Republic and Portugal.

But what will this achieve? According to Noss, the kinds of knowledge people will need in the 21st century are different from what they have needed in the past. He believes we ought to be aware of the systems that affect our lives. “There’s a very simple choice - we can either be happy for everybody to merely consume technology, or we can try to equip them with ways of understanding and controlling that technology.”

To give children the chance to develop their own games and thereby gain an understanding of how they work, the researchers are working with six to eight-year-olds in five Cambridge schools and one in London using animatedprogramming software called ToonTalk.

The program, developed by Professor Ken Kahn of Massachusetts Institute of Technology, allows children to develop their own sophisticated computer games without using text. The children are given elements that are combined to make a game, which they then play. Others - from their own school and in schools in the rest of Britain, as well as Portugal and Sweden - play it and suggest improvements. The developer then has to think about how to make those changes using ToonTalk. The researchers’ work has demonstrated that eight-year-olds are, contrary to popular belief, capable of handling complex rule systems and understanding causality. One pupil called Mitchell realised that, by altering his game so that his opponent’s character could only move up and down, but not sideways, he could win more easily.

Although it is impossible to prove, Noss says it is plausible that when Mitchell is 15 and has to understand concepts such as “why balls travel in parabolas”, he might, thanks to all this early lateral thinking, be able to extrapolate from “some knowledge hidden in his mind somewhere” the answer to such complex problems.

Celia Hoyles, who is also a professor of mathematics education at the Institute of Education and the project’s other co-director, says they have found, in the six months since the research began, that boys and girls design very different games and change them in different ways.

Girls found a simple tennis-like game that involved batting a ball uninteresting so they changed it to an underwater environment where fish were eaten by a shark.

The second phase of the project, starting this month, will use Logotron’s Openlogo, a powerful new version of Logo, which is the educational programming language created by MIT academic Seymour Papert, to allow a comparison to be made.

Playground website www.ioe.ac.ukplaygroundToonTalk www.toontalk.comLogotron www.logo.com