Lately, Keith Devlin spent much time sharing his ideas and experience about how to design video games that support good math learning.

This is not an easy task and anyone who wishes to create math video games should read Delvin´s blog.

Soon we will release our first smart game about algebra and equations, and we hope we designed a game that supports good math learning.

As a mathematics teacher, the creation of this game has been a source of constant suprise and continuous learning.

And I can tell one thing. To create a game that supports good math learning is not easy.

But it is not impossible.

Yes, we have to make allowances for playability test and engagement, and in addition, check what is learned.

But if mathematics is in the game play (that means no quizz based game), you can virtually visualize how players are thinking, and even more interesting, part of how they are learning. And you can make statistics about it.

And here you open the Pandora box, because players are VERY different. Something can work with one player, but not another. A rule can be infered correctly at once, or you can see a player like Benny, creating her own representation (not correct) of the patterns she observes and experience.

To see this broad spectrum of differences is confusing. And it is very hard to accept it because it makes things really complicated.

What to do with it?

We can change the level design as well as the game design. But it will be anyhow impossible to satisfy all.

We can also gently help the players who are not on the right track. It can be an automatic instruction shown when the game detects the player has not the right model.

But I think it is sub optimal. Intuitively, I think it would be much more efficient and meaningsful if help comes from other players. And there the social setting is important.

Playability, engagement, learning can´t be dissociated from how, when and where the game will be played. These elements have an enormous contribution in the efficiency of the game and on the learning process.

That means creating a game that supports good math learning requires another kind of test: how the setting in which the game is used influences learning and playability.

More on that in another post…