Our proposed work seeks to understand how neural activity gives rise to the accurate perception of 3D surface orientation. While it is established that the brain's computations allow 2D retinal projections to give rise to the perception of a rich 3D visual scene, how the brain does this is unknown. Here we propose a 2 pronged approach where we record neurons in 2 brain areas likely to be involved in 3D surface orientation (3D-SO) perception (CIP and V3A) and use reversible inactivation to demonstrate the likely causal role(s) of these areas. Recent work has indicated that CIP and V3A contain cells which are tuned for 3D-SO, and our proposed research will seek to determine how neural activity in CIP and V3A relates to behavior when animals are asked to perform a perceptual task judging 3D surface orientation (Aim 1). We will record from neurons in each brain area while a rhesus monkey performs a fine slant discrimination task, judging whether slanted planes are leaning backward or forward. Choice probability and ROC analysis will be used to determine the degree to which an individual neuron predicts choice behavior in the task. Population decoding and simulations will then be used to assess the differences in CIP and V3A's choice related activity at the population level. After studying the relationship between neural activity in CIP and V3A and behavior, we will using reversible inactivation to determine the relative importance of each of these areas in the ability to judge slant based on both binocular and monocular cues (Aim 2). If either area has a causal role in slant perception, inactivation should diminish performance in our psychophysical task. Our research will answer questions regarding the perception of 3D-SO, but will also provide a more general understanding of how neural activity gives rise to behavior, a question which is fundamental to neuroscience. Answering this question is critical to learning how the brain works when it is functioning normally, but will also be necessary to treat a variety of brain disorders, particularly those for which Brain-Machine interfaces offer some hope for recovery (e.g. blindness and spinal cord injury).
Determining how populations of neurons give rise to perceptual abilities is a major question in systems neuroscience; while our proposed research is focused on the underlying neural processes involved in 3d spatial orientation perception, the findings will have relevance for understanding neural coding in general. In terms of the health relevance, gaining the knowledge of how neural activity gives rise to behavior will be informative for the development of neural prostheses for the visually impaired, as well as other types of brain-machine interfaces, such as for motor impairments resulting from spinal cord injuries. Our project's focus on the perception of 3D spatial orientation also has relevance for disorders causing spatial disorientation, like Alzheimer's disease.
