Perceptual Closure

Perceptual closure describes the amazing ability of our brain to preceive or recognize objects (such as the face above)  although visual information is incomplete or corrupted by noise.


 Neural Information Dynamics

Information theoretic quantities separate and measure key elements of computation in neural systems, such as the storage, transfer, and modification of information. This way, they help to better understand the computational algorithm implemented in a neural system.  

 Predictive Coding

Brains sample sensory information to inform behavior that allows the organism’s survival in a complex environment. However, sufficient sampling of sensory information is not always possible. Sampling may be limited by the necessity of fast responses, or by unavailability of sensory information. The resulting incomplete sensory evidence often makes it impossible to recognize objects based on sensory evidence alone. Thus, neural systems have evolved to incorporate learned regularities in the environment to form predictions that help perception. To date, the mechanisms underlying the integration learned information and sensory input  in perception are still unknown. Our group works on resolving this puzzle.