Probabilistic inference based on different types of information

Our environment contains a rich spectrum of physical signals carrying different types of information. These signals can artificially be assigned to three entities: the What, relating to the type of event, the Where, pertaining to an object's position in space, and the When, giving the event a position in time. Human sensory systems transform these physical signals, such as light and sound waves, to neural representations upon which the brain performs inferences about the external causal sources of its input as well as forming predictions about future states of its environment. This basic aspect of human cognition can be observed in every-day activities such as a game of ping pong but also in specialized tasks requiring highly-trained experts, e.g. musicians identifying complex harmonies on first hearing. Probability is argued to be a key parameter in the computations involved. However, the interaction between the different types of information and their respective stochastic structure is not well understood. This project aims to describe how the information category What and the dimensions Where and When relate to event probability to drive human behavior and its underlying neural implementations.