{"id":"a85daab1-bae6-4884-8736-76b84fb02b73","arxiv_id":"1908.07302","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"In a lattice gas of run-and-tumble particles, the tagged particle keeps a non-Gaussian, roughly Laplace displacement distribution even after many tumbles, unlike the run-and-tumble polymer where long-time motion becomes Gaussian.","lead":"This paper studies how a single tagged particle moves when surrounded by interacting run-and-tumble particles, using one exactly solved polymer model and one simulated lattice gas. It finds non-Gaussian motion that, in the lattice model, persists into the diffusive regime with an approximately Laplace displacement distribution, attributed to a changing local environment.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Kurtosis has not plateaued by t≈125τ; the abstract's 'persists' is unsupported if a second crossover to Gaussian occurs.","rationale":"The reader's weakest_assumption identifies the same load-bearing point as my independent read: the lattice-model non-Gaussianity is demonstrated only over a finite simulated window, and the kurtosis is still falling at the end of that window. The paper's own caveat explicitly leaves open the possibility of a second crossover to Gaussian behavior, so the abstract's unqualified 'persists' overstates what is shown. The body is more careful, saying 'all time scales simulated' in the conclusions. Because the missing piece is exactly the asymptotic plateau, longer simulations or a scaling argument are the natural arbiter. I agree with the reader's conditional verdict: the finite-time result is valuable and likely correct, but the advertised persistence should be treated as conditional until the long-time limit is checked. No adjustment to the reader's verdict is needed.","tokens_in":14322,"tokens_out":5212,"duration_ms":58264,"concrete_test":"Extend the L=40, ρ=0.25, α=0 lattice simulations to at least t=1000τ (about 200,000 time units) with 25,000 or more histories, computing κ(t) in time blocks. If κ(t)−3 decays to zero or continues to decrease, the asymptotic non-Gaussian claim fails and the result is a finite-time transient. If κ(t) flattens at a value significantly above 3 with error bars excluding 3, the concern is resolved. Repeat with L=80 to check that the plateau, if any, is not a finite-size artifact.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim—that non-Gaussian, Laplace-like displacement persists in the lattice model—rests on the assumption that the late-time window shown in Figs. 8 and 10 is representative of the asymptotic regime. The data do not establish a plateau: at ρ=0.25 the kurtosis is still decreasing at the last simulated time, reaching κ≈6.8 at t≈125τ, and the authors explicitly write in Sec. 3, after Fig. 8, that 'It is an interesting question whether this regime is the stationary one or whether at much longer time scales a second crossover to a Gaussian diffusive regime occurs.' The abstract nevertheless drops the 'simulated' qualifier and states that non-Gaussianity 'persists', a claim that would be false if a second crossover exists. Additionally, the variance is shown only up to about 50τ while the kurtosis and distribution are shown to about 100–125τ, so the diffusive characterization of the window used for the Laplace fit is less secure than for the low-density case. No error bars are provided, so the decreasing trend and the κ≈6.8 value are not quantified against statistical uncertainty.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the motion of a tagged particle in two non-equilibrium systems of interacting run-and-tumble particles (RTPs). The first is a bead-spring polymer whose monomers perform run-and-tumble motion subject to thermal noise; for this model the authors derive an exact expression for the monomer mean-squared displacement using Rouse modes, and they identify ballistic, superdiffusive, subdiffusive, and diffusive regimes as functions of temperature and time. The second is a two-dimensional lattice exclusion process (a variant of the persistent exclusion process), studied by simulation at zero temperature, for which the authors report anomalous exponents for the tagged-particle variance, non-Gaussian displacement distributions, and, at intermediate density, an approximately Laplace-distributed displacement at long times. The non-Gaussianity in the lattice model is attributed to a dynamically changing local environment, which the authors argue is essential for deviations from Gaussianity, in contrast to the polymer case where the environment is fixed and Gaussianity is recovered after the persistence time.","tokens_in":14525,"tokens_out":3423,"duration_ms":37607,"significance":"The exact polymer result in Appendices A-C is a genuine strength: the mode-coupling calculation is self-contained, the initial condition is stated, and the resulting MSD reproduces the expected crossover structure. If the lattice-model claim is correct, the paper would provide a minimal microscopic active system producing long-lived non-Gaussian, exponential-tailed diffusion, connecting to the diffusing-diffusivity picture and to experimental observations of non-Gaussian displacement distributions. However, the lattice part of the paper is currently not at the same standard as the polymer part: the central claim that non-Gaussianity 'persists' is based on a simulation window in which the kurtosis is still decreasing, the fitted exponents are given without error bars or fitting details, and the Laplace characterization rests on a visual tail comparison. The significance of the work is therefore conditional on additional numerical evidence or on a more cautious interpretation of the simulated late-time regime.","major_comments":[{"comment":"The abstract's claim that non-Gaussianity 'persists' in the lattice model is not established by the presented data. At ρ=0.25 the kurtosis is still decreasing at the last simulated time (κ≈6.8 at t≈125τ), and the text itself explicitly allows that 'at much longer time scales a second crossover to a Gaussian diffusive regime' may occur. Since the central new claim depends on this late-time window being representative of the asymptotic regime, the authors should either extend the simulations to demonstrate a plateau or reformulate the abstract and conclusions to describe a long-lived non-Gaussian transient rather than a persistent regime.","section":"Sec. 3, Fig. 8"},{"comment":"The exponents 1.74, 1.54, and 0.81 are read from log-log slopes but no error bars, fitting procedure, or fitted time windows are given. Moreover, for ρ=0.25 the variance is shown only up to about 50τ, while the kurtosis and distribution are shown to 100-125τ, so the diffusive characterization of the window used for the Laplace fit is less secure than for the low-density case. The subdiffusive exponent 0.81 in particular needs a quantitative justification, including the range over which it is measured and its statistical uncertainty.","section":"Sec. 3, Fig. 7"},{"comment":"The evidence for the claim that the displacement is 'well approximated by a Laplace distribution' is a dotted line drawn through the large-|X| tail of ln P(X) at t=20000, together with a kurtosis of about 6.8 that is above the Laplace value of 6. A tail-slope comparison is not a distribution fit; the authors should provide a quantitative comparison, such as fit parameters, residuals, or a test against alternative exponential-tailed distributions, before asserting Laplace behavior.","section":"Sec. 3, Fig. 10"}],"minor_comments":[{"comment":"The notation Cp(t,t) is easy to confuse with the mode coefficient c_p^n; consider writing C_p(t,t) with an explicit subscript to distinguish the mode index.","section":"Eqs. (8) and (B.8)"},{"comment":"The phrase 'for t small one has t>t^2' should be rephrased, for example as 'for t<1 the linear thermal term dominates the quadratic ballistic term'.","section":"Sec. 2.1"},{"comment":"For the right panel, the dotted line with slope 0.81 is displayed over a very short interval; adding markers or shaded bands to indicate the fitted range would make the claimed exponent verifiable.","section":"Fig. 7 caption"},{"comment":"The caption should state that t=20000 corresponds to 100τ and should list the density and α values used for the simulation.","section":"Fig. 10 caption"}],"recommendation":"major_revision","confidential_remarks":"The polymer part is solid and could be published on its own, but the lattice-model claim is currently overstated relative to the evidence. I would ask the authors to either supply longer-time simulations and statistical error bars or reframe the central claim as a long-lived transient regime before acceptance."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The real news in this paper is the exact mean-squared displacement for a Rouse chain driven by run-and-tumble forces. That calculation, including the t^{3/2} superdiffusive regime for τN < t < τ, is new relative to the cited active-polymer literature, which mostly used Ornstein-Uhlenbeck or active-Brownian forces. The appendices check out: the mode decomposition is standard, the correlation functions are consistent, and the asymptotic exponents follow from the quoted integrals. I have no quarrel with the polymer half. It is a clean piece of work and a fair extension of the Rouse toolbox.\n\nThe lattice half is more fragile. The simulations at ρ=0.25 do show long-lived non-Gaussianity: kurtosis around 6.8 at t≈125τ and roughly linear log-tails, well approximated by a Laplace distribution. That is worth reporting. But the paper itself flags that the asymptotic status is open: a second crossover to Gaussian diffusion at longer times is explicitly left as a question. Given that the kurtosis is still decreasing at the last simulated time, the abstract's unqualified \"persists\" is not supported. I also note that the variance is shown only to about 50τ while the kurtosis and distribution go to 100–125τ, so the diffusive characterization of the Laplace-fit window is less secure than it looks. There are no error bars anywhere in the lattice results, so the decreasing trend and κ≈6.8 are unquantified.\n\nThe environment-fluctuation explanation is a plausible heuristic, not a derivation. The authors argue that cluster formation and dissolution create a time-dependent diffusivity, and the analogy to diffusing-diffusivity models is reasonable. But it remains an interpretation of the observed distributions, not a mechanism established from the model. That is fine as a claim, provided it is labeled as such.\n\nIn proportion: the polymer result is solid and the lattice simulations are suggestive and honestly reported. The main fix is linguistic — qualify the persistence claim with \"on the timescales simulated\" — plus error bars and, ideally, longer runs to see whether kurtosis plateaus or keeps falling. This is not a paper with a load-bearing flaw; it is a paper with an overstatement in the abstract.\n\nI would send this to a serious referee. The Rouse–RTP calculation alone justifies referee time, and the lattice observations, once properly qualified, are worth putting on record. I would cite the polymer result in work on active polymers. For the reading group, maybe: it is a useful example of how exact calculations and simulations can be combined, and it raises a clean question about asymptotics in lattice active matter.","headline":"The exact Rouse polymer result is a solid, citable derivation, but the lattice-model claim of persistent non-Gaussianity is stronger than the simulation evidence supports.","tokens_in":15037,"tokens_out":993,"would_cite":true,"duration_ms":12790,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A tagged particle in a lattice gas of run-and-tumble particles diffuses with persistent non-Gaussian, Laplace-like tails.","keywords":["run-and-tumble particles","active matter","tagged-particle diffusion","non-Gaussian diffusion","Laplace distribution","persistent exclusion process","diffusing diffusivity","Rouse model"],"falsifier":"Extend the lattice-model simulations ($\\rho=0.25$, $T=0$) to times much longer than $125\\tau$, on larger lattices, and measure the kurtosis of the tagged-particle displacement. If the kurtosis approaches 3 and the displacement distribution becomes parabolic in $\\ln P(X)$, the persistent non-Gaussianity is a finite-time transient.","tokens_in":14127,"feed_emoji":"🦠","tokens_out":7770,"duration_ms":71918,"temperature":0.7,"pith_summary":"The paper asks whether a tagged particle moving through a bath of interacting run-and-tumble particles reproduces the anomalous, non-Gaussian diffusion seen in complex environments. It studies two models: an exactly solvable polymer whose monomers run and tumble, and a two-dimensional lattice gas of run-and-tumble particles with exclusion. In both, the tagged particle's displacement is anomalous and non-Gaussian for times shorter than the persistence time $\\tau$ of the active motion. The polymer becomes Gaussian and diffusive at later times, but in the lattice model non-Gaussianity persists over the whole simulated window: at intermediate density ($\\rho=0.25$, $T=0$) the kurtosis is still about 6.8 near $t\\approx 125\\tau$, and the tails of the displacement distribution follow $\\ln P(X)\\sim -|X|$, i.e. a Laplace distribution. The paper attributes the persistence to a dynamically changing environment, in which the tagged particle alternates between free runs and being trapped by clusters, and argues this fluctuating environment is essential for non-Gaussian diffusion.","feed_headline":"A simple active lattice gas yields diffusive motion with exponential tails","feed_subtitle":"At intermediate density, tagged-particle kurtosis stays near 6.8 after 125 persistence times—far from the Gaussian 3.","key_machinery":"The load-bearing object is the displacement distribution $P(X)$ of a tagged particle, measured through its variance, kurtosis, and tail shape. In the polymer model the calculation is carried by Rouse normal modes, which diagonalize the harmonic monomer couplings and yield an exact mean-squared displacement with four time regimes; this shows when non-Gaussianity is transient. In the lattice model, the mechanism is the persistent exclusion process, a run-and-tumble lattice gas with exclusion, in which cluster formation and dissolution make the tagged particle's environment fluctuate; the paper connects this directly to the diffusing-diffusivity mechanism, arguing that a stochastically varying effective diffusivity produces Laplace-distributed displacements.","core_discovery":"The central discovery is that a minimal microscopic active system, the persistent exclusion process, can produce diffusion that is normal in the variance but non-Gaussian in shape for all simulated times, not just for short transients. Concretely, for a tagged particle at intermediate density $\\rho=0.25$ and zero temperature, the displacement variance grows linearly at late times while the kurtosis remains far above the Gaussian value 3, reaching about 6.8 at $t\\approx 125\\tau$, and for not-too-small displacements $\\ln P(X)$ grows roughly linearly in $|X|$, the signature of a Laplace distribution. The paper also shows, in the exactly solvable polymer model, that active monomers give ballistic, superdiffusive, subdiffusive, and diffusive regimes with non-Gaussian statistics only for $t<\\tau$; the late-time Gaussianity of the polymer is traced to the fact that the tagged monomer's environment does not fluctuate. The lattice-model result is interpreted through the diffusing-diffusivity mechanism: the local environment changes as clusters form and dissolve, so the tagged particle's effective diffusivity fluctuates, producing exponential tails.","pith_inferences":["If the non-Gaussian regime is truly stationary, the diffusing-diffusivity description suggests a concrete continuum limit: the tagged-particle probability density should obey a Fokker-Planck equation with a state-dependent diffusivity, and the local waiting-time distribution should match the cluster-size distribution. This is not tested in the paper.","A decisive test of the persistence claim is to push simulations beyond $125\\tau$, with larger lattices or faster algorithms: if the kurtosis returns to 3, the Laplace regime is an intermediate asymptotics and the paper's main claim would be weakened to a long transient. The authors themselves flag this open question.","The polymer result points to a general criterion for persistent non-Gaussianity in active systems: it requires a fluctuating environment for the tagged degree of freedom. One could test this by comparing tagged monomers in polymers with heterogeneously active segments, where the environment is fixed, against tracers in phase-separating active suspensions, where it fluctuates."],"forward_implications":["At intermediate densities, the persistent exclusion process provides a concrete microscopic realization of diffusing-diffusivity: exponential tails arise without introducing a stochastic diffusivity by hand.","A tagged monomer in a Rouse chain with run-and-tumble monomers is non-Gaussian only below the persistence time; chain connectivity alone does not sustain non-Gaussianity, because the monomer's environment is fixed by the chain.","For low density, the tag's early-time displacement has a three-peak structure reflecting particles running right, left, or up/down; this structure washes out into diffusive motion with non-Gaussian tails.","The cluster-size distribution of the persistent exclusion process, a power law times an exponential, gives a quantitative route from environment fluctuations to the Laplace tail of tagged-particle displacement.","The regime of diffusive but non-Gaussian motion is observable in simulations for at least two orders of magnitude in time beyond the persistence time, so it is not a short transient."],"supporting_citations":[{"why":"Defines the persistent exclusion process used as the lattice model and supplies the cluster-size distribution (power law times exponential) on which the fluctuating-environment argument rests.","marker":"[34]"},{"why":"Provides the diffusing-diffusivity model that the paper invokes to explain Laplace-distributed displacements.","marker":"[20]"},{"why":"Introduces the diffusing-diffusivity mechanism (non-Gaussian yet Brownian diffusion) that the lattice result is interpreted through.","marker":"[59]"},{"why":"Reports experimental subdiffusion with Laplace-distributed displacements in E. coli, the comparison that motivates the Laplace fit.","marker":"[58]"},{"why":"Defines the Rouse model whose normal-mode decomposition the exactly solvable polymer calculation uses.","marker":"[35]"},{"why":"Gives the single-particle run-and-tumble displacement distribution that the paper extends to interacting systems.","marker":"[33]"}],"fun_headline_variants":["Diffusive yet non-Gaussian: tagged active particles show exponential tails","Normal variance, Laplace tails: dynamics of tagged run-and-tumble particles","Active lattice gas: kurtosis 6.8, far from Gaussian 3, even at late times","Exponential displacement tails in a simple interacting active particle model","Tagged particle in active gas: diffusion is normal, shape is not"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The claim stands on the assumption that the observed non-Gaussian regime up to 125 persistence times is the real long-time limit and not a long transient before a return to Gaussian diffusion.","fun_headline_variants_meta":{"raw":{"variants":["Diffusive yet non-Gaussian: tagged active particles show exponential tails","Normal variance, Laplace tails: dynamics of tagged run-and-tumble particles","Active lattice gas: kurtosis 6.8, far from Gaussian 3, even at late times","Exponential displacement tails in a simple interacting active particle model","Tagged particle in active gas: diffusion is normal, shape is not"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000715,"raw_usage":{"total_tokens":3204,"prompt_tokens":928,"completion_tokens":2276,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":544,"completion_tokens_details":{"reasoning_tokens":2176}},"tokens_in":544,"tokens_out":2276,"duration_ms":15641,"temperature":1.0,"reasoning_tokens":2176,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T12:21:19.419574+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Extend the lattice-model simulations ($\\rho=0.25$, $T=0$) to times much longer than $125\\tau$, on larger lattices, and measure the kurtosis of the tagged-particle displacement. If the kurtosis approaches 3 and the displacement distribution becomes parabolic in $\\ln P(X)$, the persistent non-Gaussianity is a finite-time transient.","supporting_citations":[{"cited_title":"Soto and R","cited_arxiv_id":null,"evidence_quote":"Defines the persistent exclusion process used as the lattice model and supplies the cluster-size distribution (power law times exponential) on which the fluctuating-environment argument rests."},{"cited_title":"Chehkin, F","cited_arxiv_id":null,"evidence_quote":"Provides the diffusing-diffusivity model that the paper invokes to explain Laplace-distributed displacements."},{"cited_title":"and Slater G.W., Phys","cited_arxiv_id":null,"evidence_quote":"Introduces the diffusing-diffusivity mechanism (non-Gaussian yet Brownian diffusion) that the lattice result is interpreted through."},{"cited_title":"Lampo, S","cited_arxiv_id":null,"evidence_quote":"Reports experimental subdiffusion with Laplace-distributed displacements in E. coli, the comparison that motivates the Laplace fit."},{"cited_title":"Rouse, J","cited_arxiv_id":null,"evidence_quote":"Defines the Rouse model whose normal-mode decomposition the exactly solvable polymer calculation uses."},{"cited_title":"Malakar et al","cited_arxiv_id":null,"evidence_quote":"Gives the single-particle run-and-tumble displacement distribution that the paper extends to interacting systems."}],"review_version":1}