DragonBox: gaming children into learning algebra

Jul 2, 2012
Vote on Hacker News

DragonBox by We Want to KnowThe claim that a game could teach basic algebra, symbol manipulation, and equation solving in a few hours, while being truly fun and engaging, sounds just like another inflated marketing pitch. But play DragonBox, by We Want To Know, and this suddenly seems like an understatement.

Stop bitching, start a revolution

The universal challenge of teaching maths is that students need to learn both how the rules work and why they exist, often only in abstract mathematical terms rather than in easy to visualize terms, which compounds the difficulty. This means that students who cannot do everything in parallel start falling behind, thinking they are “bad at maths” and end up giving up. This is something that bothers students, parents, and teachers alike, all the more because failing at maths often means failing at scientific learning.

This is a problem that can and should be solved. DragonBox is a first step towards revolutionizing maths learning, and a very smart one at that. As clichéd as “revolutionary” sounds, when a game intended for 8yo kids proves to be interesting and effective both for 5yo kids and teenagers in high-school, it’s easy to see that something rare is happening. In Norway, where DragonBox launched and quickly reached 15 000 downloads – a very decent success for a paying app in a small market – players and press alike gave raving reviews of the game. It has so far launched on the appstore, and is being ported to Mac and PC.

Clearing the way

First and foremost, the mental roadblock of being “bad at maths” is something that DragonBox overcomes very well, by fusing learning and playing so closely that the pleasure of playing effectively becomes the pleasure of learning. This pleasure, incidentally, is what most of those who were good at maths felt, as to them the pleasure of learning was very much the pleasure of playing with maths puzzles, not to mention the pride in solving them perfectly.

Furthermore, DragonBox aims at giving its players the understanding of how equation solving works – leaving the why for another game. This means that players get deep-seated and almost instinctive understanding of algebra, through fun symbol manipulation gameplay. This primes them for learning the why part much more easily. By making players both interested and able, DragonBox is a very solid stepping-stone towards greater understanding of maths.

Play-learning recipe

The first part of the game makes players learn gameplay rules, which are algebra rules in disguise, so that they don’t even realize they do maths, but rather think they are playing a puzzle game. In the second main part, the learning symbols used at first (box, animals, and so on), are progressively replaced by mathematical signs and symbols (x instead of a box, letter constants instead of animals, etc). This offers additional challenge and fun to the players, while transferring the rules learned in the context of pure play to the context of mathematics.

Making algebra rules instinctive and fun in a few hours to pretty much any player is no small feat. The game designers, industry veteran and cognitive science PhD Patrick Marchal, and math teacher Jean-Baptiste Huyn, had to carefully balance the difficulty in order to make progression smooth and engaging. They achieved smoothness in the progression of the game by designing short levels. They kept the engagement alive by adding just one new gameplay challenge per level, making playing – and thus learning – clear and intuitive. Extensive playtesting, within and outside of classrooms, was key to balancing difficulty.

Elegant rewards and design

The challenge and rewards system DragonBox currently implements is an easy to pass, but tricky to fully win approach. It enables players to win and pass a level easily with 1 star, while earning all 3 stars per level provides a more robust challenge. The 2 additional stars, rather than just being an arbitrary point or time-based representation of success, can be won if there is the right number of cards at the end (implicitly meaning that the equation has been fully simplified), and if there was the right number of moves (meaning that the simplifying of the equation itself was done in the lowest possible number of operations). This reward mechanism encourages players not only to learn how to manipulate equations correctly, but to master the finer sequencing of algebraic operations involved.

Furthermore, the game is sequenced in chapters, each containing 20+ levels, with additional bonus levels, and plenty of progression for players to remain engaged until they complete the game. One reward beyond the stars is that each chapter is a dragon, and for each completed level, the dragon grows, which provides an effective and satisfying vizualisation of progress. UX/artistic design is one other area where the game does very well, with a very friendly universe and interface, which brings warmth to the experience. The excellent review in Wired’s geekdad shows just how well all the design elements converge and end up creating an experience in which even the 5 year old daughter of the writer is engaged by DragonBox and learns how to solve equations, even competing with her 8 year old sister to play as you can see in the video below.

The Road ahead

DragonBox is a first step for this promising team. Updates for the game are in the pipeline, and We Want To Know has announced new math games designs. I saw some exploratory prototypes in optics, electricity, and even quantum physics. Whether these end up making the cut of what they deem is worth developing is another question, but coming up with creative ideas for difficult problems doesn’t seem to be an issue here.

Scientific learning could use much more innovation. As even proven scientific notions like climate change or evolution get challenged in our societies, it certainly wouldn’t hurt for science to become easy and fun to learn for more people.