Suppose that you’re interested in how false beliefs – ‘fake news’ – spread across a population. The problem is extraordinarily complicated. What people come to believe depends on a myriad of factors: which news outlets they read; who their friends and colleagues are; their ability to distinguish between fact and fiction; their social media echo chamber; and so on. A very good way of investigating a complicated phenomenon is to model it. This involves the construction of a simplified version of it and the investigation of the behaviour of that simplified version in order to draw conclusions about the actual system of interest. Both of these steps require a curious and playful outlook.