Adding the second 50 Hz response cut late voltage test error by 28% in 960 mouse cortical connections with separate donor folds and an asinh score

We tested whether the second response to 50 Hz stimulation helps to predict later transmission along a detected cortical connection. We used Allen Synaptic Physiology r2.1 small. The main sample had 960 mouse connections from 396 inferred donors. Each donor stayed in one outer validation fold. Nested ridge tests compared cell labels, distance, age, project, and the first response with the same predictors plus the second response. The late target was a fitted voltage summary for pulses 6 to 8. An asinh score with a 0.1 mV scale gave a 28.4% error reduction when the second response was added. A paired donor bootstrap gave a central 95% range of 20.4 to 36.0%. This range holds the fitted predictions fixed. Fits at two other voltage scales and in raw mV also gave lower error. A separate human fit gave a 47.8% reduction in 111 connections from 32 donors. Early and late summaries can share accepted pulse trains and fitting assumptions. The result does not establish prediction of new independent trials or identify a release mechanism. The package preserves code, source hashes, selection, donor groups, folds, all predictions, bootstrap draws, figures, and repeat instructions. A full replay reproduced all 32 numerical output files byte for byte.

synaptic transmission · short-term plasticity · cortical connectivity · Allen Synaptic Physiology · grouped validation · burst response