Conditional learning increases the inhibition of layer 4 excitatory neurons by somatostatin- and parvalbumin-expressing interneurons in the barrel cortex

Synaptic strength underlies information processing, learning, and memory storage, yet little is known about how learning impacts synaptic inputs and outputs of specific GABAergic interneurons, particularly in the somatosensory cortex. Using a simple conditional model of learning in mice, where whisker stimulation was paired with a tail shock, we investigated plastic changes of inhibition mediated by somatostatin- (SST-), parvalbumin- (PV-), and vasoactive intestinal polypeptide-expressing interneurons (VIP-INs) in the barrel cortex. In vitro patch-clamp recordings and optogenetics revealed that conditional learning increases SST-IN and PV-IN synaptic outputs onto layer 4 (L4) excitatory neurons. A small fraction of L4 excitatory neurons in the barrel cortex receives inhibition from local VIP-INs, but learning does not affect this inhibition. Additionally, learning does not alter excitatory inputs to all three interneuron types. These findings suggest that enhanced inhibition by SST-INs and PV-INs may improve information processing and memory coding in L4 of the barrel cortex.