SteerMEM · Steering the visual content of long-term memory in monkeys

The overlap between the neural substrates and processes underlying perception and memory makes it difficult to dissociate them, and the complexity of high-level visual processing further complicates this task. By leveraging large-scale electrophysiology, intracortical microstimulation, and artificial intelligence, I will disentangle the complex relationship between visual perception and memory in monkeys, and develop innovative techniques for steering the visual content of memories and dissociate sensations from perception.First, I will record and manipulate the neural activity of the inferotemporal cortex (IT) and perirhinal cortex (PrC) in macaque monkeys to provide a comprehensive and systematic investigation of the neural mechanisms underlying visual perception and memory. Then, I will provide causal evidence on the role of conscious perception in shaping the visual content of memories, and the importance of the IT-PrC circuitry in this process. Furthermore, I will determine the threshold between visual perception and memory, testing the hypothesis that a subsets of neurons in PrC can elicit a sense of familiarity independently of any visual input. Finally, I will demonstrate how memory gates visual information to control behavior in a closed loop experiment, employing state-of-the-art wirelles recording and microstimulation in freely behaving monkeys.SteerMEM has the potential to revolutionize our understanding of visual perception and memory, paving the way to significant advancements for both neuroscience and clinics. SteerMEM will test a groundbreaking approach to manipulate long-term visual memory in a controlled and quantifiable fashion, and test the long-term applicability of this approach in ecological conditions. Finally, SteerMEM will lay the ground for AI-powered microstimulaiton in cognitive cortical prostheses for the potential treatment of memory impairments, anxiety and post-traumatic stress disorder.