【Topic】Neural circuit mechanism for inference-based flexible decision-making

　　【Speaker】Dr. Ninglong Xu Institute of Neuroscience，Center for Excellence in Brain Science and Intelligence Technology，Chinese Academy of Sciences 

　　【Time】 13:00-14:30, Oct. 19, 2021

　　【Venue】Lui Che Woo Building B101, Peking University

　　【Host】  Dr. Jianing Yu

　　Abstract

　　Making flexible decisions based on knowledge about causal environmental structures is a hallmark of goal-directed cognition in mammalian brains. Although several association brain regions including the orbitofrontal cortex (OFC) have been implicated, the precise neuronal circuit mechanism underlying knowledge-based flexible decision-making remains largely unknown. Here we investigate the role of a top-down circuit linking OFC and auditory cortex (ACx) in an auditory-guided flexible decision-making task in mouse. We established an inference-based auditory categorization task, where mice performed within-session flexible stimulus re-categorization based on changing task rules. Mice showed behavioral signature of inference, and exhibited high flexibility by rapidly switching categorization boundary upon rule changes. Using a reinforcement learning algorithm with belief state we modeled the process of knowledge-based task state inference to quantify the hidden variables associated task knowledge. By combining two-photon population imaging and projection-specific optogenetics, we found that auditory cortex neurons encoded the hidden task-rule variable, which required the feedback input from OFC. Silencing the OFC-ACx input specifically disrupted the behavioral flexibility without influencing sensory discrimination. Consistently, direct imaging from OFC axons within ACx revealed task state-related value signals, supporting the knowledge-based updating mechanism. Our data reveal a specific cortical circuit mechanism underlying inference-based flexible decision-making.