Neuronal correlates of goal-directed sensorimotor transformation in sensory, motor and higher-order cortical areas

An important function of the brain is to interpret incoming sensory information from the outside world to guide adaptive behavior. Understanding how and where sensory information is transformed into motor commands in a context- and learning-dependent manner is a key question in neuroscience. Which cortical areas are responsible for sensory perception, decision-making and motor execution is still a matter of discussion. During my Ph.D. I attempted to disentangle the coding of sensory versus decision and motor information by neurons in different brain regions along the transformation path of whisker sensory stimuli to licking motor output during goal-directed behaviors in mice. To address this question I combined electrophysiological recording of a large population of neurons across multiple brain areas with fine monitoring of behavior and movements, as well as time-resolved optogenetic manipulations.I first explored the sequence of cortical activity involved in transforming a brief whisker stimulus into delayed licking in mice. With coauthors, we discovered two crucial features related to activity evoked by whisker deflection. Firstly, an enhanced excitation of the secondary whisker motor cortex suggested its important role in connecting whisker sensory processing to lick motor planning. Secondly, a transient reduction of activity in the orofacial sensorimotor cortex helped to suppress premature licking. During the delay period, we observed widespread cortical activity that largely correlated with anticipatory movements. However, when accounting for these movements, we identified sustained activity in the frontal cortex that was essential for licking in the response period. Our results highlight key cortical nodes involved in motor planning. We next categorized neurons as regular or fast spiking, which we validated by optotagging method. We investigated how their activity changes before and after mice learn a task. We observed opposite changes in the whisker-evoked activity of regular versus fast spiking neurons in primary and secondary whisker motor cortices, but similar changes in primary and secondary orofacial motor cortices. Hence, altered excitation and inhibition in local circuits, combined with changes in long-range synaptic inputs may underlie delayed sensory-to-motor transformation. Finally, I recorded and systematically compared the neuronal activity of the cortical sensory input area (primary somatosensory cortex), a presumably decision area (medial prefrontal cortex), and the motor output area (tongue and jaw primary motor cortex) in the psychophysical whisker detection task. We have uncovered that the representation of sensory information, decision-making, and motor action is not a discrete function of each individual brain region, but rather distributed and encoded across all three regions. We proposed that the decision to lick after the whisker stimulus is a gradual process that involves the flow of information from wS1 to mPFC and tjM1. Importantly, we also observed that signal processing failures can occur at any level of this information flow. These findings highlight the complexity and distributed nature of the neural processes underlying decision-making and sensorimotor transformation. In this thesis, I have presented various findings that add to the existing knowledge and shed light on the neural circuits and computations involved in sensorimotor transformation in the mammalian brain.