The role of macaque V1 neurons in spatiochromatic processing and behavior

Abstract

Vision is critical for survival. We can easily identify objects, guide actions, and avoid collisions if our eyes are open but these abilities are severely impaired if our eyes are closed. This tremendous feat of vision appears simplistic but is implemented by complex biological processes performed by the eye and brain. Therefore, a central goal in visual neuroscience is to understand how neurons in the brain represent scenes, and how the neural activity in turn helps guide behavior. Scenes are composed of spatial and chromatic variations, herein referred to as spatiochromatic variations. In the primary visual cortex (V1) of macaque monkeys, some neurons jointly analyze edges and color, making them an ideal substrate for understanding human spatiochromatic vision. Double-opponent (DO) cells in V1 respond strongly to adjacently placed lights of opposite color and weakly to spatially uniform light of one color. These properties make them well suited for processing of color across space. However, we do not know precisely what information DO cells represent and how. Understanding how DO cells function will advance the field of visual neuroscience in three ways. First, it will help us understand how DO cells are connected to other neurons, thereby shedding light on the organization of cells in V1. Second, it will help link the neuronal responses to behavioral phenomena in color vision. Third, it will advance mathematical models of visual processing that will guide research in other fields. How information about scenes is used for behavior is incomplete without understanding the link between neural activity and behavior. A mechanistic understanding of how V1 neural activity impacts visual perception will be important for understanding the role of V1 in diseases and designing brain-machine interfaces. Using a combination of electrophysiological measurements, monkey behavior and state-of-the-art techniques, I investigated the role of V1 DO cells in the spatiochromatic processing of light, and the role of V1 neural activity in visual perception. I compared my findings about DO cells to simple cells—the best understood functional cell type in V1 that represent oriented luminance edges in scenes, and integrate signals across space roughly linearly. I pursued my research in the form of three different projects, and I report the key findings from each of the projects below. In project 1, I investigated the representation of edges by DO cells. I found that DO cells represent chromatic edges the same way as simple cells represent luminance edges. In project 2, I investigated how DO cells integrate color signals across space. I found that DO cells integrate spatial signals as linearly as simple cells meaning that both these classes of neurons simply weigh and sum the incoming light to generate a spiking response. In interpreting this result, it is important to realize that linearity is not the default mode of visual neurons but rather implies a specialized wiring. My results suggest that the specialized wiring creates linear luminance edge detectors and chromatic edge detectors in V1. Together, the results from project 1 and project 2 suggest that DO cells are similar to simple cells in many ways, and these classes of neurons have a similar mechanism of processing edges than previously thought. This property has major implications in understanding the neural circuitry of these cell classes and their contributions to image processing, which I discuss in Chapters 2 & 3. In project 3, I investigated the impact of silencing neural activity on behavior by pioneering a fast and powerful neural inactivation technique in monkey cortex. The advantage of this technique is that the neural inactivation can be reversed on a trial-by-trial basis, which was difficult to achieve previously. Inactivation of V1 led to reduced sensitivity for visual detection by monkeys suggesting that V1 neural activity impacts visual perception. This result opens doors for possible therapeutic treatments of visual impairments and investigations of many outstanding questions in the domain of perception, action and cognition, which I discuss in Chapter 4. Collectively, my research has made important strides in the field of visual neuroscience by advancing our understanding of spatiochromatic processing by DO cells, and the impact of V1 neural activity on visual perception.