To constrain and test models of V1, we need to comprehensively characterize the physiological responses of neurons as well as their main inputs in all six cortical layers. However, imaging neuronal cell bodies or axons deep in the cortex has remained a major technical challenge. The Ji lab made critical breakthroughs in applying adaptive optics to correct brain-induced optical aberrations for in vivo two-photon fluorescence microscopy, so that the activity of axons, boutons, and neurons can be captured across the full depth of cortex. Using Bessel focus scanning technology, the Ji lab also demonstrated high speed (30 Hz) volumetric calcium imaging with synaptic resolution in vivo. In awake mice and under different brain states, these technologies will be used to measure neural activity of both cortical neurons and their main sensory and modulatory inputs in response to a concerted set of visual stimuli including natural movies to rigorously test model predictions. Combined with subsequent in vitro functional connectivity mapping and post hoc cell type identification, we will generate, for the first time, datasets where every neuron within V1 is described by its location, responses to known stimuli, connectivity pattern, and cell type, to guide and validate the modeling projects.
