Axon origin gave no clear common delay difference after controls in 331 mouse V1 inhibitory connections

We tested whether axon origin helps explain output delay in mouse primary visual cortex (V1). We used Allen Synaptic Physiology release r2.1. The sample had 331 inhibitory connections from 250 source neurons and 160 inferred mice. Each axon had a clear soma or dendrite origin label. Each source neuron had equal total weight. The model controlled cell class, soma distance, intended layer, slice depth, age, sex, axon loss status, and project. The common delay ratio for dendrite versus soma origin was 0.980. Its 95 percent mouse bootstrap interval was 0.897 to 1.073. Adding origin to a ridge model increased test error by 0.91 percent. The conditional interval was -0.23 to +2.25 percent. Whole mice stayed outside training and model selection. Planned checks gave similar estimates. A later geometry check also gave no prediction gain. These data give no clear support for one common delay change from this coarse structure label. The result does not show equal delays or rule out cell type effects. The labels do not measure the axon initial segment or conduction alone. Source code, source records, computed results, figures, and repeat instructions are preserved.

axon origin · inhibitory neurons · synaptic delay · mouse visual cortex · structure and function · animal validation