{"id":"94a017f5-fd85-4040-914a-6d09da4a6d20","arxiv_id":"1908.06233","paper_version":1,"verdict":"UNVERDICTED","confidence":"HIGH","novelty_score":0.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"A review article that surveys recent advances in stochastic process theory motivated by single-particle tracking experiments, presenting no new derivation, data, or method.","lead":"This paper is a review of recent progress in understanding random motion of small particles, spanning first-passage times, single-trajectory spectra, non-Gaussian diffusion, and anomalous diffusion. A generalist might read it as a compact orientation to the statistical physics tools used in modern single-particle tracking experiments.","discovery_kind":"review","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified; the review makes no new falsifiable claim and spot-checks do not reveal transcription errors that would undermine its summary role.","rationale":"The reader correctly identifies transcription accuracy as the only meaningful vulnerability of a review article, and I agree that the reliability of the summary depends on faithful citation. However, on inspection the central recalled formulas are standard and self-consistent; the one genuinely uncheckable element, unpublished reference [90], is peripheral. The review presents no new falsifiable result, so the appropriate handling remains UNVERDICTED. A confirmatory check of the least transparent equations would fully settle the residual transcription concern, but it does not change the verdict.","tokens_in":18490,"tokens_out":5698,"duration_ms":59317,"concrete_test":"Verify the two least-derivable displayed results against their sources: re-derive Eq. (5) for the partially reactive sphere from the eigenfunction expansion in [23], and simulate fractional Brownian motion to confirm Eq. (14) for γ(ω) at α = 0.5, 1, and 1.5. If both reproduce the printed forms, the transcription risk is closed.","verdict_should_be":"UNCHANGED","load_bearing_attack":"I read the paper as a review whose central claim is that it accurately summarizes selected recent developments in stochastic processes. For such a claim to fail, displayed equations or descriptions would need to misrepresent the cited results. Spot-checking Eq. (4), Eq. (11), and the asymptotic limits of Eq. (14), the forms are consistent with standard results; the Section 4 sentence about the coefficient of variation appears to omit the small-frequency limit, but the displayed formula and its three large-frequency limits are internally coherent. The only uncheckable support is unpublished reference [90], used in one comparative sentence, and it is not load-bearing for the stated claims. I find no internal inconsistency and no unsupported novel result, so no load-bearing concern lands.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This invited-style review surveys recent developments in stochastic processes, organized around four areas: full first-passage-time densities in confined domains and for partially reactive targets; single-trajectory time-averaged mean squared displacements, ageing, and ergodicity breaking; single-trajectory power spectral densities and their amplitude fluctuations; and Brownian-yet-non-Gaussian diffusion via superstatistics and diffusing diffusivity. The review frames these developments by the new experimental possibilities of superresolution microscopy and single-particle tracking, and argues that the fluctuations of single-trajectory observables carry diagnostic information that complements ensemble averages. The paper contains no new derivations or data; its contribution is the selection, synthesis, and transcription of published results, with Figures 2 and 3 reproducing experimental and simulation data from the cited literature.","tokens_in":18569,"tokens_out":13592,"duration_ms":133981,"significance":"If the summary is accurate, the review has pedagogical and practical value for experimentalists entering single-particle tracking and for theorists seeking a compact map of recent developments. Its strengths are the clear separation of ensemble and time-averaged observables, the emphasis on amplitude fluctuations as model-selection tools, and the visual documentation in Figures 2 and 3. The equations and qualitative statements I spot-checked are consistent with standard results: the Lévy-Smirnov first-passage density, the ageing factor in Eq. (11), and the diffusing-diffusivity crossover and kurtosis behaviour in Section 5 are all recognizable from the cited literature. No free parameters or fitted quantities enter the presentation, and no new falsifiable prediction is made, so the paper should be judged as a review rather than as a primary research contribution.","major_comments":[],"minor_comments":[{"comment":"The roadmap in the Introduction misstates the section order: it assigns single-trajectory power spectra to Section 3 and non-Gaussian diffusion to Section 4, whereas the actual headings are Section 3 for time-averaged MSD, Section 4 for power spectra, Section 5 for non-Gaussian diffusion, and Section 6 for conclusions. Please correct the roadmap.","section":"Introduction"},{"comment":"As rendered, the prefactor in Eq. (4) appears to be x0/(4πDt)^{3/2}, but the normalized one-dimensional Lévy-Smirnov first-passage density requires the prefactor x0/√(4πDt^3). If the original PDF contains the square root, please ensure the typeset version is unambiguous; otherwise correct the prefactor.","section":"Equation (4)"},{"comment":"The sentence \"As function of ω=fT, γ has the unique value √2 independent of the anomalous diffusion exponent α\" should be qualified as the small-frequency (ω→0) limit; in its present form it appears to contradict the three distinct large-ω limits stated immediately after Eq. (14).","section":"Section 4, paragraph after Eq. (13)"},{"comment":"The statement \"For subdiffusion, the coefficient is negative\" is ambiguous because the coefficient of variation γ is a nonnegative quantity. If the intended statement concerns the sign of the prefactor α(α−1)Kα in the noise autocorrelation written just before, please say so explicitly rather than referring to \"the coefficient\".","section":"Section 4, paragraph after Eq. (13)"},{"comment":"The caption says the kurtosis crosses over \"from the value K=9 for a one-dimensional Laplace distribution\", but a standard one-dimensional Laplace distribution has kurtosis 6. The value K=9 is consistent with a squared-Gaussian diffusivity distribution at short times. Please reconcile the wording with the actual model distribution shown in Figure 3.","section":"Section 5, around Figure 3"},{"comment":"In the sentence beginning \"Thus, from a random walk perspective\", the kernel is written as [r²(t′+t)−r²(t′)]²; the superscript on r inside the square brackets is inconsistent with Eq. (7) and should read [r(t′+t)−r(t′)]².","section":"Section 3, paragraph after Eq. (7)"},{"comment":"Reference [90] is listed as \"unpublished\". Because it is used to support a comparison in Section 5, it should be replaced by a published account or explicitly labelled as a personal communication so that the reader can assess the claim.","section":"References"}],"recommendation":"minor_revision","confidential_remarks":"The review is largely organized around the author's own recent work and that of close collaborators, with a substantial fraction of the highlighted developments citing the author's papers. This is understandable given the author's central role in these developments, but the editor may wish to encourage a more balanced citation of independent work or an explicit statement of the review's selective perspective. This does not affect my recommendation, which is based on the manuscript's evident accuracy and internal coherence after the minor corrections listed above."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This is a review article, not a research preprint, so the usual accept/revise/reject frame doesn't quite fit. If you want a quick orientation to first-passage times beyond the mean, single-trajectory power spectra, and Brownian-but-non-Gaussian diffusion, this is a serviceable one-stop summary. The historical framing from Robert Brown through Einstein and Perrin is engaging, and the core equations I spot-checked — the Levy-Smirnov first-passage density (4), the ageing factor Lambda(alpha) in (11), and the large-frequency limits for the coefficient of variation gamma in (14) — are stated correctly. Figures are explicitly credited to their original sources.\n\nWhat the paper does well: it is honest about being a summary. It does not overclaim novelty. The selection of topics is coherent, and the emphasis on fluctuations of time-averaged quantities rather than just their means is a useful methodological message for experimentalists. The diffusing diffusivity section gives a readable account of how a linear MSD can coexist with a non-Gaussian PDF, and the crossover from Laplace to Gaussian via the kurtosis is presented clearly. The stress-test conclusion holds up: I did not find an internal inconsistency or a misquoted formula that undermines the summary role.\n\nSoft spots: most of the cited developments are the author's own work, so the 'recent developments' are weighted toward Metzler and collaborators. For a balanced survey, a reader should complement this with other groups' perspectives. The one support from an unpublished source, reference [90], appears in a comparative sentence and is not load-bearing; replacing it with a published account would improve the bibliography. The text in Section 4 on gamma states the large-frequency limits but omits the small-frequency behaviour, which is slightly misleading; the displayed equation itself is fine. And, as is normal for a review, no derivations are provided, so reliability rests on the cited literature — but my spot-checks did not turn up transcription errors.\n\nWould I send it to peer review? If it lands in a journal that publishes invited reviews, yes — it is a competent, accurate expert survey, and refereeing can only improve the bibliography and the occasional loose sentence. But if the question is whether this adds a new research result, the answer is plainly no. It deserves a serious referee as a review, not as a research contribution.\n\nFor a reading group: it could be useful for a student or a researcher entering the field, but it is not something I'd assign to an expert group.","headline":"A clear, honest review that consolidates recent stochastic-process results without adding new ones; worth refereeing as a survey, not as a research contribution.","tokens_in":19091,"tokens_out":2501,"would_cite":false,"duration_ms":27251,"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":"This review argues that modern single-particle tracking demands diagnostics beyond ensemble means: full first-passage-time densities, fluctuation spectra of time-averaged displacements, and diffusing-diffusivity models that produce…","keywords":["Brownian motion","first-passage times","time-averaged mean squared displacement","single-trajectory power spectrum","diffusing diffusivity","Brownian yet non-Gaussian diffusion","anomalous diffusion","ergodicity breaking"],"falsifier":"Recompute the key formulas from first principles or simulation: simulate a scale-free continuous time random walk and test whether the ageing time-averaged mean squared displacement equals the non-aged value times the predicted ageing factor, and simulate the minimal diffusing-diffusivity equations and check whether the one-dimensional kurtosis crosses from 9 to 3 at the correlation time. If either prediction fails, the review's central benchmarks are wrong. Alternatively, measure the coefficient of variation for a known subdiffusive system at high frequency; theory says it tends to 1, so a conflicting plateau would refute the spectral classification.","tokens_in":18248,"feed_emoji":"📊","tokens_out":9034,"duration_ms":83329,"temperature":0.7,"pith_summary":"This review, aimed at a physics audience, gathers recent theoretical advances triggered by superresolution microscopy and single-particle tracking. Its core message is that ensemble means are no longer enough: reaction kinetics at the low molecular concentrations inside cells require full first-passage-time densities, and single-trajectory data should be read through their fluctuations as well as their averages. The review argues that amplitude scatter in time-averaged mean squared displacements and in single-trajectory power spectra carries process-specific signatures that distinguish subdiffusion, Brownian motion, and superdiffusion. It also argues that apparently Brownian motion with non-Gaussian displacement distributions is naturally produced by a diffusion coefficient that fluctuates along the trajectory, with a predictable crossover back to Gaussian statistics. If these claims hold, standard data-analysis practice for tracking experiments would shift from averaging many particles toward making fluctuation statistics the primary diagnostic.","feed_headline":"Fluctuations, not averages, carry the single-particle signal","feed_subtitle":"How full first-passage-time distributions and fluctuation spectra identify diffusion regimes from few trajectories.","key_machinery":"The machinery is a set of exactly solvable stochastic-process models and their fluctuation statistics. For first passage, the paper uses eigenfunction expansions of the diffusion equation in confining domains, which split reaction-time densities into geometry-controlled, process-controlled, and domain-controlled regimes. For time averages, the central objects are the time-averaged mean squared displacement, the dimensionless amplitude measuring how far an individual trajectory's time average deviates from the ensemble mean, the ergodicity-breaking parameter, and the ageing factor for scale-free continuous time random walks. For spectra, the central object is the single-trajectory power spectrum together with its amplitude distribution and coefficient of variation as a function of the product of frequency and observation time. For non-Gaussian diffusion, the load-bearing mechanism is the diffusing-diffusivity model: a coupled pair of stochastic equations in which the particle position follows white-noise driving with amplitude proportional to the square root of the time-dependent diffusivity, while that diffusivity is the square of a mean-reverting auxiliary process; a subordination formulation makes the short-time superstatistical Laplace regime and the long-time Gaussian crossover analytically tractable.","core_discovery":"The paper's central claim is that four recent lines of work form a coherent upgrade of diffusion analysis. First, for a particle diffusing in a bounded domain toward a partially reactive target, the full reaction-time density—not just its mean—has a three-regime structure: a short-time hump set by direct trajectories, an intermediate process-dependent power-law decay, and a terminal exponential shoulder set by the confining domain; lowering the target reactivity widens a plateau of nearly equiprobable reaction times, so the mean reaction time can be atypical. Second, for scale-free continuous time random walks the time-averaged mean squared displacement remains a random quantity even in the long-measurement limit, and its scatter, quantified by the amplitude distribution and the ergodicity-breaking parameter, distinguishes such non-ergodic processes from ergodic ones. Third, the single-trajectory power spectrum is proportional to the ensemble-averaged spectrum but carries a random amplitude, and its coefficient of variation tends in the high-frequency limit to the square root of 2 for superdiffusion, the square root of 5/2 for Brownian motion, and 1 for subdiffusion, allowing regime identification from few short tracks. Fourth, a minimal diffusing-diffusivity model—a particle driven by white noise whose diffusivity is the square of a mean-reverting auxiliary process—produces a linear mean squared displacement with Laplace-distributed displacements at short times and a crossover to Gaussian displacements at long times, quantified by a kurtosis drop from 9 to 3 in one dimension.","pith_inferences":["If fluctuation diagnostics are as informative as the review suggests, the same quantities could be used to infer spatial or temporal heterogeneity of the environment from a single trajectory, a step beyond the model-class identification the review emphasizes.","The diffusing-diffusivity structure is mathematically the same stochastic-volatility structure used in financial mathematics; the first-passage and spectral diagnostics reviewed here could transfer to time-series analysis of volatility and returns.","A testable extension follows directly: in any Brownian-yet-non-Gaussian experimental system, measure the kurtosis as a function of lag time; if the short-time value and crossover time do not match the diffusing-diffusivity prediction, then alternative mechanisms such as quenched-disorder heterogeneity are needed.","The three-regime reaction-time structure suggests that search-and-reaction optimization should target the most probable reaction time rather than the mean; comparing the mean and mode of measured first-reaction-time distributions could reveal whether geometry-control or reaction-control dominates."],"forward_implications":["At nanomolar concentrations, reaction kinetics should be described by the full first-reaction-time density; the mean reaction time can be unrepresentative, and both geometry-control and reaction-control set the typical times.","Amplitude scatter and the ergodicity-breaking parameter give model-selection fingerprints: different stochastic mechanisms leave distinct scatter patterns in finite-time trajectories.","Single-trajectory power spectra can classify subdiffusion, normal diffusion, and superdiffusion from few, short tracks via the coefficient of variation, with distinct high-frequency limits and quantitative agreement with telomere, agarose-gel, vacuole, and amoeba data.","The diffusing-diffusivity model explains the coexistence of a linear mean squared displacement with non-Gaussian displacement statistics, and the kurtosis crossover provides a direct experimental estimate of the diffusivity correlation time.","Bayesian and machine-learning parameter estimators, together with newer statistical tools, are positioned as the natural next step for identifying the underlying stochastic mechanism from measured time series."],"supporting_citations":[{"why":"Supplies the classic framework and universal 3/2 power-law asymptote for first-passage time densities in unbounded domains.","marker":"[14]"},{"why":"Provides the eigenfunction-series technique used to compute full first-passage time densities in hyperspherical domains.","marker":"[21]"},{"why":"Provides the reaction-time density for a partially reactive target and the separation of diffusive and reactive contributions to the mean reaction time.","marker":"[23]"},{"why":"Introduces the amplitude-scatter distribution and the ergodicity-breaking parameter for time-averaged mean squared displacements.","marker":"[35]"},{"why":"Supplies the ageing factor for the time-averaged mean squared displacement of scale-free continuous time random walks.","marker":"[38]"},{"why":"Defines the single-trajectory power spectral analysis and the random-amplitude distribution for Brownian and fractional Brownian motion.","marker":"[49]"},{"why":"Provides the coefficient-of-variation result for single-trajectory power spectra and the experimental comparisons in four systems.","marker":"[50]"},{"why":"Introduces the minimal diffusing-diffusivity model and its short-time Laplace to long-time Gaussian crossover with kurtosis characterization.","marker":"[72]"},{"why":"Originates the idea of a stochastically fluctuating diffusion coefficient along a single trajectory, the diffusing-diffusivity concept.","marker":"[74]"}],"fun_headline_variants":["Full reaction-time densities beat means for reactive targets","Single-trajectory spectra reveal diffusion type from few tracks","Full distributions, not just means, decode diffusion regimes","Mean reaction times mislead; full densities tell the tale","Ergodicity breaking and kurtosis shifts mark non-Gaussian motion"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the formulas and model predictions quoted from earlier papers—the reaction-time density, the ageing factor, the coefficient-of-variation limits, and the diffusing-diffusivity crossover—were transcribed correctly and that those models describe the single-particle-tracking experiments to which the review compares them.","fun_headline_variants_meta":{"raw":{"variants":["Full reaction-time densities beat means for reactive targets","Single-trajectory spectra reveal diffusion type from few tracks","Full distributions, not just means, decode diffusion regimes","Mean reaction times mislead; full densities tell the tale","Ergodicity breaking and kurtosis shifts mark non-Gaussian motion"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000939,"raw_usage":{"total_tokens":4012,"prompt_tokens":944,"completion_tokens":3068,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":560,"completion_tokens_details":{"reasoning_tokens":2987}},"tokens_in":560,"tokens_out":3068,"duration_ms":20825,"temperature":1.0,"reasoning_tokens":2987,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T12:51:57.713210+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Recompute the key formulas from first principles or simulation: simulate a scale-free continuous time random walk and test whether the ageing time-averaged mean squared displacement equals the non-aged value times the predicted ageing factor, and simulate the minimal diffusing-diffusivity equations and check whether the one-dimensional kurtosis crosses from 9 to 3 at the correlation time. If either prediction fails, the review's central benchmarks are wrong. Alternatively, measure the coefficient of variation for a known subdiffusive system at high frequency; theory says it tends to 1, so a conflicting plateau would refute the spectral classification.","supporting_citations":[{"cited_title":"Godec and R","cited_arxiv_id":null,"evidence_quote":"Provides the eigenfunction-series technique used to compute full first-passage time densities in hyperspherical domains."},{"cited_title":"Grebenkov, R","cited_arxiv_id":null,"evidence_quote":"Provides the reaction-time density for a partially reactive target and the separation of diffusive and reactive contributions to the mean reaction time."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Introduces the amplitude-scatter distribution and the ergodicity-breaking parameter for time-averaged mean squared displacements."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the ageing factor for the time-averaged mean squared displacement of scale-free continuous time random walks."},{"cited_title":"Krapf, E","cited_arxiv_id":null,"evidence_quote":"Defines the single-trajectory power spectral analysis and the random-amplitude distribution for Brownian and fractional Brownian motion."},{"cited_title":"Krapf, N","cited_arxiv_id":null,"evidence_quote":"Provides the coefficient-of-variation result for single-trajectory power spectra and the experimental comparisons in four systems."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Introduces the minimal diffusing-diffusivity model and its short-time Laplace to long-time Gaussian crossover with kurtosis characterization."}],"review_version":1}