The ability to seamlessly move between global map perspectives (GMPs), such as looking at a map or taking in a scene from a high vantage point, and first-person perspectives (FPPs), such as being on the ground immersed in the scene, has obvious importance for navigation and, more broadly, flexible perspective-taking. Using a multidisciplinary approach, we will investigate the neural architectures, environmental features, and spatial representations that support these transformations. We will develop neural network models that make predictions to be tested in experiments with the rodent and human, and use that experimental data to inform the model.