{"id":"9aa13198-14a3-44e1-beda-ef4ce4d913ee","arxiv_id":"2608.03906","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Reaction delays alone can rectify active swimmers in asymmetric static activity landscapes and enhance diffusion in symmetric ones, with current direction tuned by delay time.","lead":"Swimming with a delayed reaction breaks a symmetry rule that prevents directed flow in static activity landscapes. The delay acts as a tunable knob that can create, boost, or reverse a net current in several active-particle models.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 'generic' symmetry-breaking claim overreaches: the derivations are for a linear δ-kernel, and the paper's own zero-current plateau for 2d/v+<τ<(l-d)/v+ shows rectification is a resonance effect, not a generic feature of finite delay.","rationale":"The reader's conditional verdict is appropriate. The paper convincingly demonstrates, for the specific linear convolution models and piecewise-constant landscapes, that delayed speed adaptation can produce rectification and enhanced diffusion, with simulations supporting the analytical estimates. The concern I identify is not that the mechanism is absent, but that the word 'generic' in the abstract and discussion overstates the scope. The paper itself contains a finite delay interval with zero net current, and the analytical framework is limited to linear time-translation-invariant kernels, especially the δ-kernel. A conditional verdict requiring qualification of the genericity claim and either an analytic symmetry argument or numerical tests against nonlinear/broadened kernels is the right outcome. This does not overturn the core mechanism, so the reader's CONDITIONAL verdict should remain unchanged.","tokens_in":6796,"tokens_out":24116,"duration_ms":272419,"concrete_test":"Implement the 3-step landscape of Fig. 2(d) (d=l/10, v-=10^-3 v+, D=0, Ω=10^-2 l/v+) with i) a gamma-distributed kernel of mean τ=d/v+ and width σ, and ii) a saturating Hill nonlinearity v_d(t)=V_max C(t)^n/(K^n+C(t)^n) applied to the convolution C(t)=∫Γ(t-t')v(r(t'))dt'. Scan σ and n/K and measure the stationary current. If the current survives only for σ→0 and n=1, the 'generic' claim fails outside the linear δ-kernel limit; if it persists across these variations, the concern is resolved.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that delayed motility 'generically' breaks the no-current symmetry is not supported by the paper's own results. The analytical derivations, including Eq. (6), the current-reversal intervals, and the zero-current intervals, are restricted to the linear, time-translation-invariant convolution in Eq. (3), with the main noiseless results obtained for a δ-kernel. Exponential memory and inertial cases are only simulated and show reduced currents and no current reversals. More tellingly, the text states that in the noiseless trap limit, for 2d/v+ < τ < (l-d)/v+ (with d<l/3), ascending and descending swimmers spend equal time in the v+ region, so the net current vanishes exactly for RTPs and numerically for ABPs (Fig. 2d). Rectification is therefore not a generic consequence of finite delay but a resonance phenomenon that requires the delay time to match traversal times across landscape features. If the speed response is nonlinear or state-dependent—outside the class defined by Eq. (3)—no symmetry argument is given to guarantee persistence of the current. The headline claim should be qualified to 'for linear memory kernels and resonant delay times' unless a broader argument is supplied.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript studies active particles (RTPs, ABPs, and inertial RTPs) whose propulsion speed is a delayed functional of the local activity field along their trajectory, as written in Eqs. (3) and (4). It argues that delayed motility breaks the no-current constraint for static asymmetric periodic activity landscapes, producing directed currents that can be tuned and even reversed by the delay time, and that memory enhances effective diffusion in symmetric landscapes. The main analytical results are limit-cycle estimates for noiseless, discrete-delay, step-like activity profiles (Table I and Eq. (6)); Brownian-dynamics simulations for RTPs and ABPs, plus exponential-memory and inertial variants, support the qualitative behavior but show reduced currents and no current reversals in those broader cases. The paper positions the mechanism as a generic, experimentally accessible route to self-steering and active-matter circuitry.","tokens_in":7205,"tokens_out":7421,"duration_ms":72447,"significance":"The proposed mechanism is physically interesting and, within the discrete-delay resonance regime, the paper gives transparent, parameter-free predictions (optimum leapfrog delay tau = d/v_+, maximum forward current v_+/2 in an ideal noiseless limit, and a zero-current interval) that are directly testable in feedback-controlled or photosensitive swimmers. The comparison across RTP, ABP, exponential-memory, and inertial variants is a strength, as is the explicit demonstration that translational diffusion and tumbling wash out the resonances. However, the 'generic' framing exceeds what is derived: no argument covers nonlinear or state-dependent delayed kinesis, and the paper's own results identify the effect as resonant rather than universal. No code or data are provided, so reproducibility rests on the equations and simulation details described in the text and in the (currently placeholder) Supplemental Material.","major_comments":[{"comment":"The abstract's claim that delayed motility 'generically' breaks the no-current symmetry is not supported by the paper's own results. Eq. (6) and the current-reversal intervals are derived for the linear, time-translation-invariant kernel in Eq. (3), mostly with a delta-kernel and noiseless step profiles. The text near Fig. 2(d) states that for 2d/v_+ < tau < (l-d)/v_+ (with d<l/3) ascending and descending swimmers spend equal time in the v_+ region, so the net current vanishes exactly for RTPs and numerically for ABPs. Exponential-memory and inertial cases show reduced currents and missing current reversals. Thus rectification is a resonance effect requiring the delay to match traversal times, not a generic consequence of finite delay; no argument is supplied for nonlinear or state-dependent delayed kinesis. The headline should be qualified (e.g., 'for linear memory kernels and resonant","section":"Abstract; 'Rectified transport in one dimension' (after Eq. (6), Fig. 2)"},{"comment":"The quantitative content of the Letter -- Table I, Eq. (6), the v_up/v_down expressions, and the analytical curves in Figs. 1(d) and 2 -- is derived in the Supplemental Material, which is referenced only as 'Ref. [8]' with placeholders 'xxx' and 'yyy'. The main text alone is not self-contained and these load-bearing formulas cannot be verified from the text. The Supplemental Material must be supplied and properly referenced before the manuscript is complete.","section":"Analytical estimates; Ref. [8]"},{"comment":"The abstract and introduction promise that current reversals are 'conveniently tuned via the delay time', but the text immediately before Fig. 2(f) states that there are 'missing current reversals in the inertial and exponential-memory cases'. Reversals are only demonstrated for discrete delays with a delta-kernel; for exponential memory and inertial RTPs the simulations show no reversal. This limitation should be stated explicitly in the abstract and conclusions rather than left to a later caveat.","section":"Abstract and final paragraph of 'Rectified transport in one dimension'"}],"minor_comments":[{"comment":"The caption says 'nominal speeds v_- and v_+ = v_-/2', which conflicts with the text's definition v_- < v_+. Please check the notation and make the high-/low-speed labeling consistent.","section":"Fig. 1 caption"},{"comment":"The first row appears as 'ID > ...'; this is likely a typo for 'I: D > ...'. Also, 'thermal diffusion negligible' in regime I seems opposite to the stated inequality D > v_+^2/(2 Omega), where translational diffusion dominates; please clarify.","section":"Table I"},{"comment":"The notation v[r(t), n(t)] and the integral expression v[r(t')] n(t') are confusing. Please explicitly define 'prior memory' and distinguish the inertial memory from the delayed speed functional in Eq. (3).","section":"Eq. (4) and surrounding text"},{"comment":"Reference [8] is a placeholder ('xxx', 'yyy') and Ref. [10] lacks year/pages. The Data Availability statement says no software supports the manuscript; given the extensive simulations, a public repository or at least a detailed simulation protocol would aid reproducibility.","section":"References and Data Availability"}],"recommendation":"major_revision","confidential_remarks":"The manuscript fits the journal and the discrete-delay resonant mechanism is a solid, testable contribution. The main revision needed is to align the abstract and conclusions with the demonstrated scope (linear memory kernels, resonant delay times) and to include the Supplemental Material with the derivations. No concerns about citation ethics; Ref. [15] appears closely related and is cited."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nYou should know two things about this paper. First, the core mechanism is real and new: a discrete delay in the speed response lets RTPs and ABPs hop across low-activity traps and, in a three-step asymmetric profile, that produces a directed current, with sign tunable by the delay. The simulations support the story, and the leapfrog picture is physically transparent. Second, the paper sells it as \"generic,\" but its own equations show the current is a resonance effect: for a whole interval of delay times the net current vanishes exactly. So the generic claim is too strong as stated.\n\nWhat's actually good: the rectification bound in Eq. (6), the regime table for diffusion, and the comparison across four model classes (discrete delay, exponential memory, ABP, inertial RTP). The simulations are extensive—10^9 to 5x10^9 time steps, many realizations—and the analytical estimates for the step profile match them closely. No free parameters fitted to the data, which is a plus.\n\nWhere it's soft: the genericity claim. All analytical results come from the linear convolution in Eq. (3), and most from the δ-kernel. The paper's own zero-current interval 2d/v+ < τ < (l-d)/v+ shows that rectification requires the delay distance to match landscape features. Exponential memory and inertia blunt the effect and kill the reversals. So the abstract's \"generically breaks this symmetry\" overreaches; the honest version is \"for linear memory and resonant delay times.\" Also, the derivations are in the SI, so the main text leans on simulations for the broader cases. No data/code release is a minor issue, not a fatal one.\n\nThe stress-test note about the resonance plateau holds up; it's not a manufactured flaw.\n\nBottom line: this deserves a serious referee. The mechanism is novel enough to merit attention, and the simulations are careful. But the authors need to either supply a symmetry argument for nonlinear/state-dependent memory or qualify the claim. I'd read it as a strong Letter about a specific delayed-feedback ratchet, not a universal symmetry-breaking principle.","headline":"A genuinely new rectification mechanism, well simulated, but the \"generic\" claim is a resonance effect in disguise; still worth a serious referee.","tokens_in":7507,"tokens_out":2245,"would_cite":false,"duration_ms":21493,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Delayed speed adaptation generically breaks the symmetry against current rectification in static activity landscapes, producing tunable directed transport and enhanced diffusion.","keywords":["time-delayed motility","kinesis","reaction delay","activity landscapes","microswimmers","run-and-tumble particles","active Brownian particles","rectification"],"falsifier":"Take a one-dimensional periodic three-step activity profile with speeds v+, v0, v- and low-region width d, and a swimmer whose speed responds to the local activity with delay tau. The paper predicts a peak forward current near tau=d/v+ and a backward current for (2-v0/v+)d/v+ < tau < 2d/v+ when d<l/3 and v- is small. A programmable microswimmer (e.g., feedback-controlled Janus particle) reproducing this setup should show both features; if either is absent under low-noise conditions, the symmetry-breaking claim fails.","tokens_in":6767,"feed_emoji":"⏱️","tokens_out":9359,"duration_ms":90771,"temperature":0.7,"pith_summary":"The paper seeks to establish that a finite response delay in how a swimmer sets its speed can turn a static pattern of faster and slower regions into a ratchet. In ordinary, instantaneously responding active particles, a symmetry condition called local detailed balance forbids any net current from such a static landscape. With delayed speed adaptation, that symmetry is generically broken: in an asymmetric periodic activity profile the particle drifts in one direction, and in a symmetric profile its spread is enhanced. The current can be tuned in magnitude and sign by the delay time, even reversed, and the effect appears for run-and-tumble particles, active Brownian particles, and inertial active particles alike, without walls, gradients, potential forces, or translational diffusion. If true, delayed kinesis is a generic, experimentally accessible steering mechanism for autonomous active matter.","feed_headline":"Reaction delays turn static activity patterns into one-way streets","feed_subtitle":"Tune the response delay and the same activity ramp drives swimmers forward, backward, or not at all.","key_machinery":"The load-bearing object is the delayed speed functional v[r(t)] = integral_{-infinity}^t Gamma(t-t') v(r(t')) dt', in which the speed at time t is a memory-weighted average of the local activity along the trajectory. For exponential memory the kernel is Gamma(s)=tau^{-1} exp(-s/tau); for the analytically tractable discrete delay it is Gamma(s)=delta(s-tau). The discrete kernel produces the 'leapfrog-and-tumble resonance'—the delay is matched to the traversal time of the slow interval, so the particle crosses a trap before the low speed is felt—and this resonance is what breaks local detailed balance, sets the magnitude and sign of the rectified current, and explains the peak of the enhanced","core_discovery":"The central claim is that a delayed speed functional—the current speed is a causal average of the activity the particle experienced along its earlier trajectory—breaks the local detailed-balance constraint that otherwise prevents rectification in static activity landscapes. For a discrete delay, the mechanism is a leapfrog resonance: when the delay tau equals the time d/v+ a swimmer needs to cross a low-activity interval at full speed, it can pass the slow region before the slow speed \"kicks in\" on the way up a three-step asymmetric profile, but gets trapped on the way down. The resulting stationary current has the upper bound v+/2/(1+tau v+/l) in the optimal limit, approaches about 20% of v","pith_inferences":["Inference: a purely passive spatial modulation of light, fuel, or chemoattractant could act as a programmable conveyor in microfluidics, with the delay time as the dial and no external forcing.","Inference: the predicted reversal windows give a sharp experimental test—sweeping tau in a fixed three-step landscape should show forward current near tau=d/v+, backward current in the interval (2-v0/v+)d/v+ < tau < 2d/v+, then zero—so absence of both features would undercut the claim.","Inference: the linear convolution model is a minimal case; analogous symmetry breaking likely occurs for nonlinear or state-dependent speed memories that retain finite causal memory, though the quantitative current estimates would need to be rederived.","Inference: for organisms, delayed kinesis may serve as a navigational strategy distinct from taxis—symmetric landscapes would boost dispersal, asymmetric ones would give direction, even without gradient sensing."],"forward_implications":["In a static, asymmetric periodic speed profile, a finite response delay produces a net directed current, whereas the same profile with instantaneous speed response produces none.","The delay time is a control knob: tuning it can maximize forward transport, reverse the current, or shut it off, without changing the landscape.","In symmetric profiles the delay mechanism enhances the effective long-time diffusivity, with an optimal boost at the leapfrog delay; noise progressively washes out the resonance.","The generality across run-and-tumble, active Brownian, and inertial active particles means no walls, potential interactions, imposed gradients, or translational diffusion are needed to see the effect.","For real microswimmers with biochemical or feedback delays, delayed kinesis should generically produce systematic motion in heterogeneous environments and offers a route to programmable 'soft active circuitry'."],"supporting_citations":[{"why":"States the local detailed-balance restriction on transport in static activity landscapes, which the paper's delayed motility is claimed to break.","marker":"[7]"},{"why":"Introduces the delayed active swimmer in a velocity landscape, the direct model from which the speed-memory functional is adapted.","marker":"[4]"},{"why":"Reports sensory-delay transport in active colloids, a recent result generalized here to arbitrary delay kernels and inertial particles.","marker":"[15]"},{"why":"Demonstrates engineering sensorial delay to control phototaxis, establishing the experimental feasibility of tunable delayed speed response.","marker":"[2]"},{"why":"Provides experimental evidence of density modulations in active colloids controlled by sensory delays, anchoring the effect in real systems.","marker":"[5]"},{"why":"Supplies the equivalence between run-and-tumble and active Brownian particles used to argue the results apply to both.","marker":"[6]"},{"why":"Shows an autonomous active Brownian ratchet in two dimensions without delay, the contrast for the one-dimensional delayed mechanism.","marker":"[9]"},{"why":"Provides the stochastic time-delay formalism that justifies the discrete delta-memory kernel as a limit with infinitely many hidden degrees of freedom.","marker":"[13]"}],"fun_headline_variants":["Delay tunes swimmers: forward, backward, or stalled","Leapfrog resonance: delayed reaction flips active particle flow","Reaction delay turns static activity ramps into one-way tracks","One delay knob controls direction and speed of active matter","Delayed kinesis gives self-steering without walls or gradients"],"cache_read_input_tokens":2688,"weakest_assumption_plain":"The result assumes the delayed speed is a linear, time-translation-invariant average of the activity the particle previously encountered; if real speed responses are nonlinear or depend on orientation or internal state beyond that history, the generic symmetry-breaking mechanism need not survive.","fun_headline_variants_meta":{"raw":{"variants":["Delay tunes swimmers: forward, backward, or stalled","Leapfrog resonance: delayed reaction flips active particle flow","Reaction delay turns static activity ramps into one-way tracks","One delay knob controls direction and speed of active matter","Delayed kinesis gives self-steering without walls or gradients"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000154,"raw_usage":{"total_tokens":997,"prompt_tokens":643,"completion_tokens":354,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":387,"completion_tokens_details":{"reasoning_tokens":270}},"tokens_in":387,"tokens_out":354,"duration_ms":4171,"temperature":1.0,"reasoning_tokens":270,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-05T05:55:22.828552+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a one-dimensional periodic three-step activity profile with speeds v+, v0, v- and low-region width d, and a swimmer whose speed responds to the local activity with delay tau. The paper predicts a peak forward current near tau=d/v+ and a backward current for (2-v0/v+)d/v+ < tau < 2d/v+ when d<l/3 and v- is small. A programmable microswimmer (e.g., feedback-controlled Janus particle) reproducing this setup should show both features; if either is absent under low-noise conditions, the symmetry-breaking claim fails.","supporting_citations":[{"cited_title":"Metzger, S","cited_arxiv_id":null,"evidence_quote":"States the local detailed-balance restriction on transport in static activity landscapes, which the paper's delayed motility is claimed to break."},{"cited_title":"Universal transport of active colloids with sensory delay in motility landscapes","cited_arxiv_id":"2604.25745","evidence_quote":"Reports sensory-delay transport in active colloids, a recent result generalized here to arbitrary delay kernels and inertial particles."},{"cited_title":"T¨ opfer, M","cited_arxiv_id":null,"evidence_quote":"Provides experimental evidence of density modulations in active colloids controlled by sensory delays, anchoring the effect in real systems."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the equivalence between run-and-tumble and active Brownian particles used to argue the results apply to both."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Shows an autonomous active Brownian ratchet in two dimensions without delay, the contrast for the one-dimensional delayed mechanism."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the stochastic time-delay formalism that justifies the discrete delta-memory kernel as a limit with infinitely many hidden degrees of freedom."}],"review_version":1}