STN: depolarization-block recovery missing — voltage stays pinned at ~−15 mV after high-stim step

[JCorson]

Observation

When the STN preset is driven with a sustained suprathreshold step (e.g. +5 µA/cm² × 200 ms in the F-I sweep), the cell enters depolarization block mid-step and fails to repolarize after the step releases. Voltage hangs at ≈ −15 mV across the entire post-stimulus window with no escape.

Reproduce: open the UI, select Subthalamic Nucleus, run the F-I Curve protocol with the default 0–5 µA/cm² range, and inspect the +5 µA/cm² trace. Same behaviour at lower amplitudes once the cell crosses into block.

Doublet bursts at 181 Hz intra-burst frequency in the same regime are also a partial-depolarization-block artifact — STN tonic firing under sustained depolarisation should remain regular (Bevan & Wilson 1999).

Root cause

INaP (Magistretti & Alonso 1999 kinetics, make_inap_channel) is implemented with only an activation gate p, no inactivation. At V ≈ −15 mV the cell sits on a stable depolarized attractor:

• INaP fully open → steady inward current.
• Fast Na inactivation (h) saturated low → g_Na contributes nothing.
• K conductances are open but balance against INaP at the depolarized plateau.

The model has no escape mechanism. Real STN cells recover via slow Na⁺ inactivation, ATP-dependent K conductances, or extracellular K⁺ accumulation, none of which are modelled.

Scope

This is not STN-specific — any preset that uses INaP (cortical pyramidal #311, FSI #301, dopaminergic #318, STN #305) is potentially exposed. STN surfaces the limitation most clearly because the F-I sweep extends to +5 µA/cm² where it reliably enters block.

Options

1. Add slow Na inactivation (s gate) to make_inap_channel, or implement a separate make_inas_channel per Magistretti & Alonso 1999 that includes their slow inactivation. Universal fix; affects all presets that use INaP.
2. Cap the STN F-I max_stimulus at ~2 µA/cm² so the default UI sweep stays out of the block regime. Workaround only.
3. Document and accept for now (the immediate path taken in fix(presets): STN tonic pacemaking and AP shape (closes #305) #319): explicit known-limitation comment in the STN preset.

Preferred direction: option 1 (universal biology fix), tracked here.

Reference

PR #319 closes #305 with the autonomous-pacemaking + AP-shape phenotype only. Depol-block recovery was deliberately deferred and noted in the STN preset comment.