{"id":"7922be02-41e9-4c58-9dcb-c849d85bd520","arxiv_id":"2607.20496","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":2.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"Action potentials are active-medium threshold pulses that annihilate on collision; solitons are conservative nonlinear-dispersive waves that collide elastically, so the two should not be conflated.","lead":"Action potentials in nerves are not solitons: they are threshold-driven pulses produced by active ion currents, while solitons are shape-preserving waves in passive nonlinear media. This essay clarifies the terminology and warns against conflating the two concepts in neuroscience.","discovery_kind":"review","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Soliton-formation timescale rests on a single iHJ parameter set; sensitivity analysis needed before drawing a blanket conclusion about nerve signals.","rationale":"The core distinction between APs (active medium, threshold, all-or-none) and solitons (conservative, nonlinear-dispersive) is correctly stated and well supported by classical references. The paper's application of this distinction to nerve signals, however, relies on a quantitative timescale argument: the iHJ model's soliton train emerges only after ~110 ms and ~2 m, exceeding AP timescales. This sub-claim is the linchpin for the conclusion that solitons are not appropriate for nerve signals. The reader's weakest_assumption identifies this exact vulnerability. Our read agrees: the simulation in Fig. 3 uses one parameter set and one initial condition, and the paper's own caveat about the conservative model's validity over long times/distances undermines confidence in the extrapolation. A parameter sweep over physiological ranges and initial amplitudes is a straightforward, decisive test. If some sweeps yield sub-millisecond soliton formation, the timescale argument fails, though the fundamental physics distinction would remain. This warrants keeping the verdict CONDITIONAL, as the reader set, with the added condition of sensitivity analysis.","tokens_in":10033,"tokens_out":6168,"duration_ms":63339,"concrete_test":"Repeat the iHJ simulation of Fig. 3 with the same parameters but vary the initial pulse amplitude A0 from 0.5x to 10x the value used in [19] (and width B0 from 0.5x to 2x), measuring the earliest time at which two distinct soliton peaks separate. If any case yields separation time below 1 ms (after converting dimensionless time with the same scaling used to obtain 110 ms), the timescale argument fails. Additionally, perform the same scan across the plausible parameter ranges for P, Q, H1, H2 reported in [19].","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's negative conclusion about solitons in nerve signals relies heavily on the claim that soliton trains in the improved Heimburg-Jackson (iHJ) model require ~110 ms and ~2 m to emerge (Section 3, Fig. 3). This quantitative argument is based on a single parameter set (P=-0.2186, Q=0.004230, H1=72.14, H2=1.000) and a single sech^2 initial condition from [19]. The paper itself acknowledges (Section 3) that other parameter combinations, even beyond physiological range, produce different behaviors (Fig. 4), and it cautions that the conservative iHJ model may not be justified over such distances and times ('Even before considering the question if all assumptions made during derivation... are justified'). Yet this timescale conclusion is used in the Abstract and Section 4 to state that 'solitons are not appropriate for describing nerve signals.' If a physiologically plausible parameter set or a larger initial excitation produces solitary waves within the millisecond timescale of an AP, the blanket statement loses force. The argument would require at least a sensitivity analysis over measured parameter ranges and initial-condition space.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper argues that action potentials (APs) and solitons are fundamentally different objects: APs are solitary waves in an active, dissipative medium generated by the Hodgkin–Huxley ion mechanism, whereas solitons are nonlinear waves in conservative dispersive media arising from a balance of nonlinearity and dispersion. The authors present definitions (Section 2), numerical examples of soliton-train formation from a microstructured-solid model and the improved Heimburg–Jackson (iHJ) model (Section 3), and examples from the Lieberstein (Hodgkin–Huxley-type) model for AP generation and collision. They conclude that the time scale for soliton emergence from an arbitrary initial excitation (~110 ms, ~2 m) exceeds that of typical APs, and that 'solitons are not appropriate for describing nerve signals.'","tokens_in":10363,"tokens_out":5884,"duration_ms":58458,"significance":"If the conclusions are accepted, the paper provides a useful terminological and conceptual clarification for neuroscience, distinguishing solitary waves from solitons in the strict mathematical-physics sense. The strength of the paper is its emphasis on the different generation mechanisms (active vs conservative) and interaction properties (annihilation vs elastic collision), supported by established references and reproducible numerical methods (pseudospectral scheme, parameters listed in the appendix). The paper is synthetic rather than novel in its derivations, which is appropriate for an essay. However, the quantitative claim about soliton emergence timescales and its use to rule out solitons in nerve signalling rests on a narrow parameter choice and is not yet supported by sensitivity analysis; this limits the force of the blanket conclusion.","major_comments":[{"comment":"The statement that soliton formation requires ~110 ms and ~2 m is based on a single simulation of the iHJ model with parameters P=−0.2186, Q=0.004230, H1=72.14, H2=1.000 (Fig. 3). The paper itself cautions (Section 3) that 'even before considering ... if all assumptions made during derivation ... are justified' and that other parameter combinations produce different behaviors (Fig. 4). Since the Abstract and Section 4 use this timescale to conclude that 'solitons are not appropriate for describing nerve signals,' the claim is load-bearing. A sensitivity analysis over the measured physiological parameter ranges and initial-condition space, or an explicit qualification that the conclusion is parameter-specific, is needed before the blanket statement can be accepted.","section":"Section 3, Fig. 3; Abstract"},{"comment":"The comparison 'time-scale of emerging solitons ... exceeds the time-scale of the usual AP existence' is not quantitatively defined. The AP is shown in Fig. 6 over tens of milliseconds and centimeters, while the soliton train in Fig. 3 is computed in dimensionless time. The paper does not state whether 'AP existence' means the duration of the AP at a fixed point, the time for the AP to traverse a typical axon, or another metric. Without a definition and a direct comparison, the reader cannot evaluate the claim. Please specify a quantitative measure and, ideally, plot both processes on comparable axes.","section":"Section 4; Abstract"},{"comment":"The final conclusion 'solitons are not appropriate for describing nerve signals' conflates two separate points: (i) APs and solitons are defined by different physical mechanisms, which is well argued; and (ii) the emergence timescale of solitons from a localised initial condition is too long to be relevant for APs, which depends on the specific model and parameters. The authors themselves note in Section 4 that van der Waals-type models [23,24] may capture AP-like phenomena better than the iHJ model. The conclusion should be separated into these components and the timescale argument qualified accordingly.","section":"Section 4, final paragraph"}],"minor_comments":[{"comment":"The phrase 'time-scale of the usual AP existence' is undefined; please specify the intended metric (e.g., AP duration at a point, propagation time over a typical axon) and give the numerical value used.","section":"Abstract / Section 4"},{"comment":"'typical dimensions of nerve axons' is vague; human axons can exceed 1 m in length, so a 2 m propagation distance is not obviously 'significantly more.' Clarify what is meant (e.g., typical axon length, AP wavelength, or conduction distance of interest).","section":"Section 3, paragraph after Fig. 3"},{"comment":"There is a typo: 'action potnetial' should be 'action potential.' Also, 'psuedospectral' appears in Section 3; should be 'pseudospectral.'","section":"Appendix, after eq. (6)"},{"comment":"The top and bottom panels have different units (space in cm vs dimensionless space, time in ms vs dimensionless time). To make the qualitative comparison clear, either use consistent variables or explicitly label the axes in each panel; the current figure may mislead readers comparing the two panels.","section":"Fig. 7"},{"comment":"The paper contrasts 'solitary wave' and 'soliton' but does not formally define 'solitary wave.' Consider adding a one-sentence definition to make the distinction precise.","section":"Section 2"},{"comment":"References [8] and [9] are dated 2026 with volume numbers; if this is a preprint, please indicate the publication status or update the references.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The numerical examples are drawn largely from the authors' own prior work ([7,8,12,18,19]), which is acceptable for a synthesis but means the paper does not contain independent verification of the key numerical claims. The main substantive issue is the overgeneralised timescale conclusion; a careful revision addressing the sensitivity would make the paper publishable."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nRead Engelbrecht, Tamm, and Peets on action potentials and solitons. The short version: it's a review essay with no new math or data, but it does a clean job of laying out the conceptual difference between an AP and a soliton and illustrating it with simulations. The punchline is that the distinction is not new—Scott made it in 1999—and the paper says so. What it adds is a set of numerical examples showing soliton-train formation in a Boussinesq-type membrane model versus AP generation in a Hodgkin-Huxley-type model, plus a collision comparison (solitons pass through, APs annihilate). Those figures are useful pedagogically.\n\nThe main conceptual claim—that APs are solitary waves in an active medium, not solitons in a conservative nonlinear dispersive system—is sound and well-supported by the definitions and the examples. I don't think anyone who knows the physics would dispute it.\n\nWhere the paper gets soft is the quantitative add-on. In Section 3 and the Abstract, the authors claim that soliton emergence from an arbitrary excitation takes roughly 110 ms and 2 m in the improved Heimburg-Jackson model, which far exceeds AP timescales, and they use this to conclude that solitons are not appropriate for describing nerve signals. That timescale comes from one parameter set (Fig. 3) and one initial condition. The paper itself acknowledges that other parameter regimes, even non-physiological ones, produce different behavior (Fig. 4), and it adds the caveat that the conservative-model assumption may not be justified over such distances and times. But the caveat doesn't stop the Abstract from stating the timescale as if it were general. A sensitivity analysis over measured parameter ranges and initial-condition space would be needed before making the blanket statement. Even if such an analysis weakened the timescale argument, the conceptual distinction would still hold—the generation mechanisms are different regardless.\n\nThe citation pattern is heavily self-referential, but that's understandable because the numerical examples are taken from the authors' earlier papers. It's not a hidden circularity; it's a review of their own body of work.\n\nWho is this for? It's for researchers in mathematical biology and neuroscience who need a clear statement of terminology, and for referees who encounter loose uses of \"soliton\" in submitted papers. It's not a primary research contribution. It deserves a serious referee mainly because the terminology debate is live and the paper's overgeneralization should be caught and fixed before publication. I'd send it to review, with the expectation of revision on the timescale claim.\n\nOverall: a useful, honest essay with one load-bearing but fixable weakness.","headline":"A clear, honest review essay that restates the AP/soliton distinction with helpful simulations but overreaches in its timescale-based conclusion.","tokens_in":10762,"tokens_out":2911,"would_cite":false,"duration_ms":30450,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["92C20","35Q51","35Q53"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper argues that action potentials and solitons are physically distinct: action potentials are threshold-driven active-medium waves, while solitons are conservative nonlinear-dispersive waves, and soliton formation from arbitrary excit","keywords":["action potential","soliton","solitary wave","Hodgkin-Huxley","Boussinesq equation","Heimburg-Jackson model","nerve signal","active medium"],"falsifier":"A direct experiment or simulation that produced a well-separated membrane soliton train from an initial pulse within less than a millisecond and a few centimeters using physiologically plausible membrane parameters would contradict the timescale argument. Alternatively, a head-on collision experiment in which two mechanical pulses in the membrane pass through each other unchanged, rather than annihilating, would weaken the claim that AP-like signals are not solitons, though it would not by itself settle the classification.","tokens_in":9985,"feed_emoji":"⚡","tokens_out":3865,"duration_ms":36499,"temperature":0.7,"pith_summary":"This paper tries to settle a terminological and conceptual dispute in neuroscience: whether the action potential can be described as a soliton. It argues that the two are different physical objects: an action potential is an active, threshold-driven solitary wave that annihilates on collision, whereas a soliton is a conservative, nonlinear-dispersive wave that passes through other solitons with only a phase shift. Using numerical examples from a Hodgkin-Huxley-type model and from Boussinesq-type equations (including the improved Heimburg-Jackson membrane model), it shows that soliton formation from an arbitrary initial pulse takes roughly 110 ms and meters of propagation, far exceeding the timescale of action potentials. The upshot is that the term 'soliton' should be reserved for its mathematical-physics meaning, and the accompanying mechanical wave in the membrane should be treated as a solitary wave, not the signal itself.","feed_headline":"Action potentials are not solitons","feed_subtitle":"Soliton formation in membrane models takes ~110 ms and meters, while nerve pulses live for milliseconds over centimeters.","key_machinery":"The argument is carried by a contrast between two mathematical descriptions. On one side is the Hodgkin-Huxley paradigm, in particular the Lieberstein modification including inductance, whose equations contain an ion-current source term and a threshold: subthreshold inputs decay, superthreshold inputs produce the same action potential regardless of amplitude. On the other side is the Boussinesq paradigm, specifically the improved Heimburg-Jackson equation U_TT = C0^2 U_XX + P U U_XX + Q U^2 U_XX + P U_X^2 + 2Q U U_X^2 - H1 U_XXXX + H2 U_XXTT, a conservative equation with nonlinear and dispersive terms that supports analytic soliton solutions of the form U(xi) = 6(c^2 - C0^2)/P / (1 + sqrt(1","core_discovery":"The central claim is that the physics of emergence of action potentials and solitons is fundamentally different. Action potentials are generated by voltage-gated ion channels in an active medium: an above-threshold stimulus triggers a stereotyped asymmetric pulse whose amplitude is independent of stimulus strength, and two counter-propagating action potentials annihilate each other. Solitons are solutions of conservative nonlinear dispersive equations like the KdV or Boussinesq equations, where the balance of nonlinearity and dispersion produces shape-preserving waves whose speed depends on amplitude and which survive collisions with only a phase shift. The paper demonstrates numerically tha","pith_inferences":["If the ~110 ms / ~2 m soliton-formation timescale holds across physiologically plausible parameters, then in short axons the mechanical wave accompanying an AP is better viewed as a forced response to the AP rather than an independent propagating signal.","The comparison suggests a testable extension: systematic parameter sweeps of the iHJ/Boussinesq model under physiological constraints could map the region where soliton formation is fast enough to matter; if such a region exists, the blanket conclusion would need refinement.","The paper's reasoning implies that van der Waals type models, which capture annihilation on head-on collision, may be more appropriate for mechanical aspects of the AP than the original Heimburg-Jackson formulation, but the terminology point stands regardless of model choice."],"forward_implications":["The term 'soliton' should not be used as a synonym for 'solitary wave'; an action potential is a solitary wave in an active medium, not a soliton.","Mechanical waves accompanying action potentials should be interpreted as solitary waves that may be driven by the action potential, rather than as the signal itself.","Models of nerve pulse propagation must incorporate the active ion mechanism to reproduce threshold behaviour and head-on annihilation; conservative Boussinesq-type equations alone do not generate a pulse on physiological timescales.","Collision experiments provide a sharp test: head-on action potentials annihilate, while membrane mechanical pulses pass through with phase shift; any model must match this distinction.","The terminology of mathematical physics matters for interpreting numerical results: a model supporting soliton solutions does not make the biological signal a soliton."],"fun_headline_variants":["Solitons can't explain nerve pulses","Nerve signals vs solitons: different physics","Why action potentials aren't solitons","Soliton timing fails for nerve pulses","Membrane solitons are too slow for nerves"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The conclusion that solitons are irrelevant to nerve signals depends on the specific improved Heimburg-Jackson parameter set in which soliton formation takes about 110 ms and two meters; if other physiological parameter regimes or other Boussinesq-type models formed solitons on sub-millisecond timescales, the blanket dismissal would lose force.","fun_headline_variants_meta":{"raw":{"variants":["Solitons can't explain nerve pulses","Nerve signals vs solitons: different physics","Why action potentials aren't solitons","Soliton timing fails for nerve pulses","Membrane solitons are too slow for nerves"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000468,"raw_usage":{"total_tokens":2104,"prompt_tokens":617,"completion_tokens":1487,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":361,"completion_tokens_details":{"reasoning_tokens":1417}},"tokens_in":361,"tokens_out":1487,"duration_ms":11068,"temperature":1.0,"reasoning_tokens":1417,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-02T11:09:49.497812+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A direct experiment or simulation that produced a well-separated membrane soliton train from an initial pulse within less than a millisecond and a few centimeters using physiologically plausible membrane parameters would contradict the timescale argument. Alternatively, a head-on collision experiment in which two mechanical pulses in the membrane pass through each other unchanged, rather than annihilating, would weaken the claim that AP-like signals are not solitons, though it would not by itself settle the classification.","supporting_citations":[],"review_version":1}