The discovery by Hubel and Wiesel more than 60 years ago that neurons in visual cortex exhibit remarkably selective responses to specific features of visual stimuli such as the orientation of edges and their direction of motion provided a fundamental mystery of cortical circuit organization that remains unsolved to this day: How do individual neurons compute their functionally selective response properties? The answer to this question is critical to understanding cortical circuit function and development, but until recently we lacked the tools that would allow us to dissect the various factors that contribute to a given response property. By employing in vivo functional imaging and selective stimulation of excitatory and inhibitory neurons at columnar, cellular, and synaptic levels of resolution in the visual cortex of species that have well-developed columnar maps (tree shrew and ferret), we are gaining fundamental new insights into the logic used by cortical circuits to build sensory representations.

While the selective responses of neurons in primary sensory areas of cortex are well known, how these cortical neurons contribute to perceptual learning remains unclear. At one extreme, V1 neurons may act as a bank of quasi-linear filters which contribute relatively constant sensory signals to downstream areas independent of behavioral context. Alternatively, depending on the behavioral task, the responses of V1 neurons could undergo changes that enhance the sensory basis for behavioral performance. To address the contribution of V1 to perceptual learning, we are using in vivo imaging techniques to explore the responses of layer 2/3 neurons in the visual cortex of tree shrews that learn to make a fine visual discrimination.

The proper construction of cortical circuits depends on a complex sequence of events orchestrated by the interaction of molecular guidance cues and activity dependent mechanisms of synaptic plasticity. We have developed a chronic ferret in vivo imaging paradigm that has made it possible for the first time to track the development of functional cortical representations in individual animals with columnar, cellular, and synaptic level resolution. One set of experiments is probing the emergence of orientation selective responses prior to and after eye opening, and a second examines the developmental mechanisms that are responsible for integrating the inputs from the two eyes to create a single coherent binocular representation.

Visual perception is the product of synaptic interactions between neurons in a number of different cortical areas. The functional organization of visual cortical areas beyond primary visual cortex and the nature of their interactions with primary visual cortex remains poorly understood. We are employing a combination of functional imaging and anatomical tracing techniques to map the representation of visual stimuli in extra-striate visual cortical areas and to probe the functional synaptic organization of the reciprocal connections with primary visual cortex in the tree shrew and ferret. Future experiments will employ optogenetic techniques to examine the contribution of feedforward and feedback projections to the responses in different cortical areas, and to visually guided behavior. We also plan to explore the development of intercortical networks and the role of visual experience in their proper formation.