Search

The capacity of the neocortex to process sensory stimuli relies on interactions between millions of neurons connected by trillions of synaptic connections in dedicated networks. If we are to understand how the brain represents external input, we must study the principles of neuronal interaction at the network level, and this requires us to uncover how connectivity between neurons relates to their function. To investigate the connectivity-function relationship, we developed a novel experimental approach to reveal the functional specificity of local synaptic connections between different cell types in layer 2/3 (L2/3) of the mouse primary visual cortex (V1). For pyramidal cells (PCs), connection probability was related to the similarity of visually driven neuronal activity. PCs with the same preference for oriented stimuli, or responding similarly to naturalistic stimuli formed connections at higher rates than those with dissimilar visual responses. This point to the existence of fine-scale PC subnetworks dedicated to processing related sensory information. In contrast to PCs, parvalbumin expressing/fast spiking (PV/FS) interneurons received dense inputs from surrounding PCs with diverse feature selectivities. PC to PV/FS interneuron connections were an order of magnitude stronger than connections between PCs. This provides a mechanistic explanation for the broad orientation tuning of most PV/FS interneurons. On the other hand, PV/FS interneurons provided divergent outputs to surrounding PCs. To relate the patterns of synaptic connectivity to network activity, we studied how patterns of neuronal co-activation are structured by visual stimuli with different statistical features. Among PC populations, patterns of neuronal correlations were largely stimulus-dependent, indicating that their responses were not strongly dominated by the functionally biased recurrent connectivity. In contrast, visual stimulation only weakly modified co-activation patterns of PV/FS cells, consistent with the observation that these broadly tuned interneurons received very dense and strong synaptic input from nearby PCs with diverse feature selectivities.