CAREER: Multi-Area Neural Dynamics Underlying Perceptual Decision-Making

Animals and humans can flexibly adapt to changing contexts and select appropriate actions to achieve behavioral goals. This natural intelligence depends on coordinated interactions between multiple brain areas, with disruptions in this coordination linked to disorders such as autism, schizophrenia, and dementia. However, we do not yet understand how interconnected brain areas coordinate their computations to support flexible behavior. This project directly addresses this open question. Understanding how distributed computation for flexible behavior emerges in the brain will advance both artificial intelligence and translational neuroscience. Because natural intelligence remains the gold standard for AI, insights from this work will inform the development of AI systems capable of flexible reasoning under uncertainty, with potential applications in autonomous systems and self-driving vehicles. Previous systems neuroscience discoveries in motor cortex helped enable brain-computer interfaces (BCIs) now undergoing clinical trials. Extending this understanding across multiple interacting brain areas could support more sophisticated and adaptable BCIs. In addition, the project will release open-source analysis approaches to the broader community, building on the lab's track record with efforts such as WaveMAP and PhysMAP that are widely used by researchers worldwide. Finally, this CAREER award will expand interdisciplinary training opportunities for students at all levels — from high school through postdoctoral — at Boston University, thus building a more competitive STEM workforce in the United States. This CAREER project investigates how populations of neurons across four clinically relevant brain areas mediate flexible decision-making, the process of combining both sensory evidence and context to select appropriate actions to achieve behavioral goals. For example, when driving home, you might need to go left at one intersection and right at another intersection. However, the brain mechanisms supporting such flexible perceptual decisions remain poorly understood. The team will train macaque monkeys to perform a flexible decision-making task which decouples perceptual judgments from motor actions. Macaque monkeys are an exceptional model system because their brains share profound structural and functional similarities with the human brain, making findings directly relevant to human cognition and disease. To solve this task, the brain must perform an exclusive-or (XOR) computation. The researchers hypothesize that the prefrontal cortex integrates sensory and contextual information to perform the XOR computation and identify the action, while downstream premotor areas transform actions into limb movements. A second hypothesis is that coordination between brain areas involved in performance monitoring and premotor areas regulates the speed and urgency of decisions. To test these ideas, the team will combine simultaneous high-density electrophysiology with optogenetic manipulations to determine how activity in one brain area shapes computations in another. Applying optogenetic approaches in macaques will also accelerate their eventual translation to humans, opening new possibilities for circuit-based therapies for neurological and psychiatric disorders. These basic science advances will also provide key data for building multi-area recurrent neural network models of flexible behavior and biologically inspired AI systems capable of flexible decision-making under uncertainty, a priority area for NSF. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.