Skip to content

SCLapicqueLIFNeuron

Module: sc_neurocore.neurons.models.lapicque

This count-neutral project identity preserves SC-NeuroCore's historical repetitive exact-flow, hard-reset LIF recurrence. It exists so compatibility behaviour is never deleted and is never mistaken for the complete Lapicque 1907 polarization experiment.

Python
from sc_neurocore.neurons.models import SCLapicqueLIFNeuron

neuron = SCLapicqueLIFNeuron(tau=20.0, resistance=1.0, dt=1.0)
voltage, events = neuron.simulate_complete(1_000, 5.0)
assert events.sum() == 200

For constant injected current $I$,

$$ v_{n+1}=v_\infty+(v_n-v_\infty)e^{-\Delta t/\tau},\qquad v_\infty=v_\mathrm{rest}+RI. $$

If the candidate reaches v_threshold, it emits and hard-resets to v_reset. The paired sc_lapicque_lif.toml/JSON schemas, all five runtime lanes, and the existing Q16.16 sc_lapicque co-simulation/formal surfaces retain this exact contract. This identity adds no literature-model count.

For the counted source identity, source voltage input, strength-duration law, and no-reset first-attainment semantics, see LapicqueNeuron.