세미나

When sensory information is incomplete or ambiguous, the brain relies on prior expectations to infer perceptual objects. Despite the centrality of this process to perception, the neural mechanism of sensory inference is not known. Illusory contours (ICs) are key tools to study sensory inference because they contain edges or objects that are implied only by their spatial context. By recording >150,000 individual cortical neurons using a combination of mesoscale two-photon calcium imaging and multi-Neuropixels measurements, we identified a sparse subset of neurons in the primary visual cortex (V1) and higher visual areas that respond emergently to ICs. Computational analyses show that these highly selective ‘IC-encoders’ mediate the neural representation of IC inference. Strikingly, selective activation of these neurons using two-photon holographic optogenetics was sufficient to recreate IC representation in the rest of the V1 network even in the absence of any visual stimulus. These data outline a new model in which local, recurrent circuitry in the primary sensory cortex reinforces sensory inferences by selectively promoting activity patterns that match prior expectations. More generally, selective reinforcement of top-down predictions by pattern-completing recurrent circuits in lower cortical areas may constitute a key step in predictive inference.