A question-answering system needs to reason about unobserved causes to correctly answer questions of the type that people face in everyday conversations. Recent neural network models that incorporate explicit memory and attention mechanisms have taken steps towards this capability. However, these models have not been tested in scenarios that require reasoning about the unobservable mental states of other agents. We propose a new set of tasks inspired by the well-known false belief test to examine how a recent question-answering model performs in situations that require reasoning about latent mental states. We find that the model is only successful when the training and test data bear substantial similarity because it memorizes how to answer specific questions. We introduce an extension to the model that simulates the mental representations of different participants in a reasoning task, and show that this capacity increases the model's performance on our theory of mind test.