Calibration under blank light recovered constant electrical input in a leaky neuron model with matched scatter

We asked whether light-scatter corrections preserve the order of known electrical inputs in a spike-latency map. We used a leaky neuron model with fixed initial voltage, imposed neighbor voltages, and constant matched scatter. Percentage subtraction reversed a twofold gap-weight order across two matched light backgrounds. Exact matching of blank latency also reversed a 4% weight difference across backgrounds because conductance scatter changed the membrane time constant. An inverse latency curve measured under blank light recovered constant added input exactly. With eight calibration steps and six separate test amplitudes, the median absolute current error was 0.25% and the maximum was 2.53% across 7,290 chosen model settings. For a time-varying source, the inverse measured an exponential average of input before the target spike. A light mismatch remained inseparable from electrical input. The method identifies gap conductance only when source voltage is known. This is a proof and a synthetic benchmark. No animal recording or anatomical connectome was analyzed. The package preserves code, proofs, all results, source provenance, figures, and repeat instructions.

electrical synapse · gap junction · spike latency · light scatter · current calibration · functional connectivity · leaky integrate-and-fire