In 61 human cortical cell pairs, the reverse 20 Hz response ratio gave no clear prediction gain with strength, subtype and geometry controls

We tested whether the measured response ratio in one direction of a two-way excitatory connection helps predict the ratio in the other direction. We used released human temporal cortex recordings at 20 Hz. The main set had 61 pairs from 18 donors. Each test donor was omitted from regression fitting and numeric preprocessing. Controls included published cell subtypes, soma geometry and both first response amplitudes. Adding the reverse log response ratio increased equal-donor mean squared log error by 0.16%. A conditional 95% donor resampling range allowed error reductions from -4.90% to +4.31%. Repeating all fits in 1,000 donor resamples gave a wider diagnostic range, from -13.65% to +15.31%. This check had a median of 12 distinct donors per draw. Penalty, subtype and signal quality checks also gave no clear gain. These results do not establish equivalent or independent responses. They limit one prediction rule for an initial response at 20 Hz in selected surgical tissue. The package preserves new code, source hashes, source row keys, model states, predictions, figures and repeat instructions. Original measurements remain at the versioned source.

human cortex · reciprocal connections · synaptic dynamics · paired-pulse ratio · donor validation · functional connectivity