Emergence of Feature-Specific Connectivity in Cortical Microcircuits in the Absence of Visual Experience

Visual response properties and connectivity of L2/3 pyramidal cells after dark rearing. A, B, Cumulative distributions of orientation selectivity index (OSI) for visually responsive neurons (A) and of the coefficient of variation (CV) of responses to the preferred grating direction (B) at P13–P15 (gray), P22–P26 (black), and after dark rearing (red, dashed line). Insets, Average and SD of means of OSI and CV of all imaged planes (number of planes: P13–P15, n = 90; P22–P26, n = 125; dark-reared, n = 48). C, Schematic showing the experimental design. D, Example sextuplet of neurons shown in a transformed in vivo image (top), in a brain slice (middle), and during whole-cell recordings (bottom). Scale bar, 30 μm. E, Membrane potential recordings from neurons shown in D. Evoked spikes in one cell and average traces of related postsynaptic potentials in the other cells are shown in each column. Traces indicating synaptic connections are colored in red. Some traces show stimulus artifacts clearly distinguishable from EPSPs. F, Polar plots showing normalized response strength to different grating directions from the cells in D and E. Synaptic connections between the neurons are indicated by red arrows. G, Overall connection probability between excitatory L2/3 neurons at eye opening, and in more mature normally and dark-reared animals (Chi-squared test). H, Relationship between connection probability and difference in preferred orientation (ΔOri) for responsive and orientation selective pairs (OSI > 0.4) (Cochran–Armitage test). P13–P15, n = 13 mice; P22–P26, n = 18 mice; dark-reared, n = 6 mice. Numbers above bars denote number of observed connections out of connections sampled.