The Mechanistic Role of Visual Areas at the Interaction of Social Behavior and Drug Effects in Nonhuman Primates - PROJECT SUMMARY To control the use of illicit drugs, we must know how they affect the brain in its entirety. Drugs such as MDMA and methamphetamine influence social emotions, targeting regions associated with emotional and reward regulation, but their effects on sensory processing remain understudied. This project aims to investigate the impact of MDMA and methamphetamine on the visual and associative regions of the macaque monkey brain using electrophysiology and deep neural networks. We will determine how social viewing affects drug-taking and integration, hypothesizing that positive social stimulation reduces drug-taking impulses and that MDMA biases prosocial free viewing. Neurons in visual cortex and pulvinar nucleus are predicted to integrate visual features, influencing visual saliency and attention. Using machine learning, we will visualize features activating neurons, including those previously considered “non-visual” or semantically uninterpretable. We predict that neurons processing local features present in conspecifics will show stronger effects during drug administration, from early visual cortex throughout associative cortex. We will develop an interpretable deep model of vision accounting for drug effects, using a novel codebase (ATHENA-N) to simulate enhanced connectivity, variability reduction, and synchronization in cortex. Long-term properties of proto-social and other neurons will be tracked in multiple visual areas using chronically implanted microelectrode arrays and naturalistic social behavioral tasks under repeated drug administration. This will assess chronic drug effects and inform future therapies. Our novel machine learning methods will reveal basic processing of social behavior in various brain regions and the extent of MDMA and methamphetamine effects on the visual cortex and subcortical structures like the pulvinar and ventral striatum. This project will demonstrate that sensory region neurons can be selectively affected by drugs based on their role in social behavior and will provide a fully interpretable mechanistic model of social visual processing, analyzing image information with and without drugs. This research will advance our understanding of social cognitive functions in early sensory cortex and their influence by drug use, highlighting the importance of sensory processing in social interactions and informing potential treatment durations for chronic drug use.
