Visual neuroscience is finally beginning to achieve a "vertically integrated" understanding of the retina, bridging all levels from molecules to microcircuits to behavior. Success could be achieved for all retinal microcircuits in just a decade, if progress were sped up drastically. Such acceleration will be attained by generating the following foundational data and disseminating it to the community. (1) We will use genetic control of ganglion cell types to pinpoint their specific roles in a suite of ethologically relevan, visually guided behaviors. Our functional explorations will be guided by the tracing of downstream pathways into subcortical and cortical regions using genetic techniques. (2) We will apply two-photon calcium imaging and serial electron microscopy to a single patch of retina to yield complete information about its activity and connectivity, i.e. "the functional connectome." This will be complemented by genetically targeted studies that correlate light and electron microscopy in separate retinas. (3) We will physiologically characterize inner retinal pathways through "projective field" measurements, i.e., massively parallel recording of population responses to direct stimulation of single bipolar and amacrine cells, employing both electrical and optical techniques. All these investigations will be linked by shared standards for typing neurons based on molecular markers, 3D structure, and functional signatures, as well as standardized visual stimuli and behavioral paradigms. We will initially develop the standards to facilitate collaboration between the members of this team. Disseminating them will enable the entire community to collectively pursue a vertically integrated approach to the retina. The ultimate impact will be transformative: a new generation of efficient coding theories and network models of retinal function solidly based on empirical data. Integrative neuroscience seeks to bridge the gap between the microscopic level of cells and the macroscopic level of behavior and mind. Such vertical integration is often professed as a goal, but has mostly been achievable only in invertebrate nervous systems. The retina now offers an unprecedented opportunity for vertically integrated neuroscience in the mammalian CNS. This approach will eventually extend to encompass the entire visual system.
Relating normal retinal microcircuits to visual function could aid those attempting to develop blindness therapies based on regenerating cells and their wiring. It could also aid the development of retinal prosthetics that directly stimulate ganglion cells and computationally emulate the bypassed microcircuitry.