A fundamental question in computational neuroscience is understanding how the brain represents and learns environmental information, from the neural level through to perception and behavior. We are building computational neural models based on the appealing hypothesis that images, movies, and sounds have predictable and quantifiable statistical regularities to which the brain is sensitive. Using vision as a paradigmatic example, we are currently particularly interested in understanding contextual processing in the brain, as a function of (1) spatial context: what surrounds a given feature or object; (2) temporal context: what has been observed in the past, i.e. adaptation; (3) context given by behaviorally relevant eye movements; (4) context given by attention and the particular task.

Additionally, a critical way to make progress is utilizing computational tools directly in experimental design and analysis. For example, we have worked extensively on spike-triggered approaches, leading to richer, non-linear characterization of neurons in retina and cortex. We are currently applying knowledge about the structure of visual scenes, to manipulate the statistics directly in experiment (in collaboration with Adam Kohn's lab, AECOM).