Faculty

Bharathi Jagadeesh
Assistant Professor (and Core Staff WNPRC)
Ph.D. Neuroscience
Northwestern University, 1993
Office phone: (206) 543-3029
bjag@u.washington.edu

Community of Science page

 

NEURAL BASIS OF VISUAL LEARNING

When we look at a scene, we see a complicated mixture of what is actually in front of our eyes, what we know, what we remember, and what we are looking for. The visually responsive areas of our brain transform the visual world around us into the neural code that is the basis of our perceptions. In primary visual cortex, the code is dominated by what is actually there, and the study of primary visual cortex has concentrated on understanding the cortical circuits that generate the basic visual code. But as visual information travels through the brain, more and more complex characteristics of the visual image are extracted, and active processes - such as learning, memory, and attention - dramatically modify the visual code.

How are complex visual scenes and images represented in the brain, and how does the representation of that information change with experience and memory? In order to find out, in our lab, we record the neural activity of cells that respond selectively to different highly complex visual images. In one experiment, we examine how the neural responses change as an observer learns to pick one object out of the scene. Our experiments suggest that the learning causes at least one change in the cortical circuit that codes these scenes: when an observer learns to pick a particular object out of a scene, neurons that respond well to other objects in the scene are suppressed. Thus, populations of cells that differentiate between complex images learn to "select" the object that is of particular importance in the task. The change in the neural responses to the scene represent a modification of the cortical circuit that codes both the images themselves and their relevance for behavior.

In our lab, we will use a variety of techniques, including single and multiple unit neurophysiological recording in behaving animals to focus on the neural circuits that code complex visual information and the modification of those circuits by learning and memory.