Review Claim by Claim

 Quick Explanation Copied

 Long Explanation

Reported results

Djioua (sole author; preprint dated August 10, 2026, zero incoming citations, no peer-review status stated) reformulates the 1952 Hodgkin–Huxley (HH) framework — classically, rapid Na+ influx followed by K+ efflux measured under voltage clamp Membrane potential and AP chapter — on two fronts: ionic conductance becomes a lognormal probability density of transmembrane ion-transit times (convolved exponentials interpreted via the central limit theorem), and the action potential (AP) becomes four overlapping membrane-potential state transitions (PSP → DEP → RPP → RRP), each a lognormal-CDF sigmoid with five parameters (τ, σ, α, t0, Ên) — 21 free parameters per 12-ms spike window Djioua 2026 HH reformulation preprint. No new experiments were performed; the study refits the original squid voltage-clamp data and lamprey reticulospinal APs from a prior in vitro study (same citation).

• HH sodium data (1952): lognormal fits yield an equivalent transient capacitance g∞ rising sigmoidally from ~5 to 20 mF/cm² over 25–75 mV depolarization and dropping abruptly above 75 mV (author: high depolarization inhibits Na+ permeability); onset t0 ≈ −0.175 ms stays constant. Potassium fits were possible but the original recordings were too short for precise parameters HH sodium/potassium lognormal refits.
• Lamprey RS fits (1,413 spikes, SNR ≥ 40 dB): E_Na ≈ 65.4 mV (author: consistent with the Nernst equation at 10 °C), E_K ≈ −89.8 mV vs the stated −90 mV theoretical value, depolarization threshold ≈ +9 mV above rest, and net ionic charge balancing to within 0.1 pmol (Table 1: PSP +0.20, DEP +0.92, RPP −1.63, RRP +0.47 pmol; computed sum −0.04 pmol) Nernst potential recovery from lamprey APs.
• Sequential coupling: onset-time distributions shift and broaden DEP→RPP→RRP (means 14.6/15.4/16.8 ms; SDs 7.16/7.24/7.28 ms); inter-transition delays of 0.8 and 1.4 ms, the latter with CV 1.38 that the author flags as a possible mixture of processes Stochastic coupling of AP transitions.

Critical assessment

• Circular validation (key caveat). The author states the decomposition "was guided by known physiological quantities, such as the Nernst potentials... and the approximate resting potential" — so the recovered E_Na/E_K partly encode their own targets, and agreement with theory is not fully independent Constraint-guided fitting admission.
• No model comparison. No RMSE/AIC or head-to-head fit against the original HH equations; the beta-to-lognormal approximation is asserted by analogy (Papoulis), not statistically tested. Classical HH remains the mechanistic baseline against which modern extensions are still benchmarked Nerve axon signal transmission review.
• Loose stochastic argument. Lognormal limits arise from multiplicative cascades; the paper bridges its additive (Erlang/convolution) and multiplicative representations via a Fenton–Wilkinson approximation with specially chosen interaction means — a construction, not a derivation (same preprint citation).
• Degeneracy and internal inconsistency. Acknowledged degeneracy (infinitely many t0/τ/σ give the same conductance peak); spike counts differ between sections (1,413 extracted vs 1,417 decomposed); no code or data availability, no funding statement; AI-assisted text/figures acknowledged; two 2025 self-citations to conference papers (same preprint citation).
• Narrow demonstrated generality. Two datasets only (1952 squid; one 2009 lamprey preparation), and the model has no voltage-dependent dynamics, so it cannot predict novel-stimulus responses without refitting — unlike dynamic HH extensions actively developed today, including unified Kv SciML models fitted to 2,969 cells Unified Kv channel SciML model, homeostatic HH-type regulation studies Channel homeostasis timescale study, and HH-plus-GNN network simulators HH-plus-GNN brain simulator.

Verdict and what would change it

The lognormal/four-transition framework is a flexible descriptive primitive whose headline validations (Nernst recovery, charge balance) are partially built in by the fitting constraints. Decisive, currently missing tests: (1) constraint-blind decompositions that still recover Nernst potentials; (2) quantified fit superiority over HH on the same squid traces; (3) selective abolition of the DEP component under Na+-channel blockade and of RPP under K+-channel blockade; (4) replication across species. Confidence in this assessment: moderate — it rests on the paper's own reported numbers, not independent verification.

Compare Fit Quality vs Original HHTest Q10 Scaling HypothesisBest Evidence: TTX/TEA Ablation Tests

Author review: Moussa Djioua

 BGPT Paper Review

Study Novelty

Recasting HH conductance as lognormal ion-transit densities and the AP as four coupled state transitions is a fresh theoretical lens, but lognormal primitives in biosignals build on the author's own kinematic-theory work and on established lognormal-neuroscience literature, making this a reformulation rather than a new mechanism.

Scientific Quality

Internally consistent mathematics, a large spike dataset (SNR ≥ 40 dB), and plausible Nernst recovery, offset by circular validation (fits guided by the very potentials it recovers), no head-to-head fit statistics, acknowledged degeneracy, a 1,413-vs-1,417 spike-count inconsistency, no code/data availability, no peer-review status, and loose CLT reasoning bridged by Fenton-Wilkinson approximation.

Study Generality

Demonstrated on only two datasets (1952 squid axon; one lamprey preparation); the model lacks voltage-dependent dynamics and cannot predict novel-stimulus responses without refitting; claimed extensions to ECG and networks remain untested.

Study Usefulness

A practical parameter-estimation pipeline from ordinary AP recordings (Nernst potentials, threshold, charge, coupling delays) would be genuinely useful for electrophysiology and spike simulation if validated; utility is contingent on constraint-blind and pharmacological tests.

Study Reproducibility

Equations and marker definitions are fully specified and inputs are published datasets, so refitting is possible in principle; however, no code, no data links, no fit-quality statistics, 21 parameters per spike, and guided decomposition hamper exact reproduction.

Explanatory Depth

Offers a mechanistic narrative (multiplicative transit times, CLT emergence, agonist-antagonist coupling, cross-scale self-similarity) deeper than pure phenomenology, but the central CLT bridge is a constructed existence argument and mechanisms are not experimentally verified.