{"id":"a1dfa278-7958-4657-ac6f-4f6c4ba562f3","arxiv_id":"1908.04119","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Quincke rollers under pulsed electric fields are shown to perform programmable run-and-tumble and Lévy walks, with collective dynamics resembling bacterial suspensions.","lead":"This paper programs tiny rolling colloids to perform run-and-tumble and Lévy random walks by pulsing an electric field. The same particles, in groups, form clusters and turbulent-like flows that look like bacterial suspensions.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Tumble-angle uniformity is not established: α≈0 and the MSD/VACF fits only constrain the first angular moment, so the 'any random walk' claim needs a direct distribution test.","rationale":"The reader's weakest assumption correctly identifies the same load-bearing concern: the persistence index α near zero only establishes that the average cosine of the tumble angle vanishes, not that the tumble-angle distribution is uniform or that successive reorientations are independent. My reading sharpens this concern by noting that the quantitative validation itself is insensitive to the higher angular moments: for exponential run times, the velocity autocorrelation decays as exp[-t(1 - ⟨cos δ⟩)/τ], so any distribution with ⟨cos δ⟩ = 0 yields the same VACF and the same MSD as uniform reorientation. Therefore the excellent agreement in Figs. 2c and 2d cannot discriminate between genuine random tumbles and a narrower, symmetric tumble-angle distribution. Direct angular statistics are needed to settle the claim. Since the reader already assigned a CONDITIONAL verdict for this reason, no verdict adjustment is needed; the condition should be stated as a required direct measurement of the tumble-angle distribution and its independence across events.","tokens_in":15887,"tokens_out":8082,"duration_ms":99660,"concrete_test":"Reanalyze the recorded trajectories for the τT/τmw = 20 condition: segment each trajectory into runs and stops, and for every off-field interval compute the signed tumble angle Δθ between the incoming run direction and the outgoing run direction. Test the empirical distribution of Δθ against Uniform(-π, π] with a Kuiper or Watson test, and also check independence of successive Δθ values and of Δθ on the incoming direction. If the non-uniformity is statistically significant while α remains near zero, the reported MSD/VACF agreement is insufficient evidence for truly random tumbles and the 'any random walk' claim must be revised; if the distribution is uniform, the concern is resolved.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing assumption is that a sufficiently long off-field interval (τT >> τmw) makes the next Quincke rotation axis uniformly random and independent of the previous run direction. The supporting evidence, α = ⟨cos Δθ⟩ ≈ 0 in Fig. 1e, constrains only the first Fourier mode of the tumble-angle distribution. Because the theoretical MSD and VACF used for validation (Eqs. D1–D5, following Refs. 32 and 62) depend on the run-vector covariance, which for a renewal random walk is governed by the mean turn-angle cosine, a non-uniform distribution with zero first moment (e.g., symmetric turns near ±π/2 with equal probability) would reproduce the same MSD and VACF as a uniform distribution. Thus the central claim that the walker emulates any random walk, and specifically that tumbles are truly random, is not established by the data shown. The phrase 'any random walk' also exceeds the demonstrated scope of two run-time distributions, but that overstatement is secondary; the unresolved angular uniformity is the prerequisite for the general claim.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports an experimental realization of a colloidal 'Quincke roller' whose motion can be programmed as a run-and-tumble or Lévy walk by applying a pulsed DC electric field. Run durations are drawn from exponential or power-law distributions and encoded as pulse widths; the field-off interval sets the tumble time and, through the Maxwell-Wagner relaxation time, the degree of memory between runs. The authors compare measured mean-squared displacement and velocity autocorrelation functions with analytic expressions from Angelani (2013) and Detcheverry (2017) using independently measured or prescribed parameters and report good agreement. They further show that dense populations of these walkers display swarming, clustering, and turbulent-like velocity fluctuations with anomalous number fluctuations, and they argue that the platform can emulate essentially any random walk.","tokens_in":16122,"tokens_out":16391,"duration_ms":161677,"significance":"The single-particle results are a significant technical advance: they provide a table-top system in which run-time statistics, speed, and turn duration are independently tunable, with negligible Brownian noise (Péclet number ~10^6). The use of measured or prescribed parameters rather than fitted ones is a strength, and the agreement with published analytic MSD/VACF expressions is convincing as far as the presented observables go. If the missing tumble-angle distribution is supplied, the platform would justify the 'any random walk' claim and could serve as a testbed for theories of active matter. The collective-dynamics observations are suggestive and connect naturally to bacterial suspensions, but they are less tightly quantified than the single-particle data.","major_comments":[{"comment":"The randomization of the tumble is supported only by the persistence index α = ⟨cos Δθ⟩ ≈ 0 (Fig. 1e). For a renewal walk with independent runs, the theoretical MSD and VACF used in the paper (Eqs. D2–D5) depend on the run-vector covariance, which is governed by the first moment ⟨cos Δθ⟩; they cannot distinguish a uniform tumble-angle distribution from any other distribution with zero first cosine moment (e.g., symmetric ±π/2 turns with equal probability). Since the abstract's claim that the strategy can 'emulate any random walk' rests on the tumble being truly random, please report the full measured distribution of Δθ and a quantitative test of uniformity (e.g., a Kolmogorov–Smirnov test against the uniform distribution), or at minimum show that the first several Fourier harmonics of the turn-angle distribution are flat.","section":"Random reorientation, Fig. 1e"},{"comment":"Equation (D5) for the Lévy-walk VACF is printed as V²/(τ+τ_T) [t0^γ/((γ−1)t^{1−γ})], which scales as t^{γ−1} and has incorrect dimensions. Differentiating the Lévy MSD in Eq. (D3) via the relation stated in the text gives V²/(τ+τ_T) [t0^γ/(γ−1)] t^{1−γ}. Please correct Eq. (D5) and verify that the theoretical curve in Fig. 2i is computed with the corrected form.","section":"Appendix D, Eq. (D5)"},{"comment":"For the Lévy walk, the MSD is only described as 'consistent with' the t^{3−γ} scaling, and no fitted exponent or uncertainty is given. Because the superdiffusive exponent is the quantitative signature of a Lévy walk, please report the measured exponent from a power-law fit over the scaling regime with a confidence interval and state the fit range. The sentence 'particle's displacement follows the desired distribution' is not supported by any displayed displacement distribution; the run-time distribution is imposed by the signal and is not an output validation. Please either add the measured flight-length or displacement distribution or remove the statement.","section":"Run-and-Tumble and Lévy walks, Fig. 2h"},{"comment":"The cluster statistics and the anomalous number-fluctuation exponents a (cited as 0.89 and 0.84) and the energy-spectrum exponent −8/3 are presented without error bars or fit details, and the cluster definition depends on a threshold distance chosen in the range 1.4d–1.6d (Appendix E). Please provide a sensitivity analysis of the reported exponents over this threshold range and report the fitting procedure and uncertainties. This is needed to support the quantitative comparisons with bacterial suspensions.","section":"Appendix E and Fig. 4"}],"minor_comments":[{"comment":"'V = 0.84 m/s' should presumably be '0.84 mm/s' to be consistent with the quoted run velocities.","section":"Fig. 2 caption"},{"comment":"The text refers to the 'R´ eclet number' immediately after defining the Péclet number; this should read 'Péclet number'.","section":"Appendix C"},{"comment":"'S2(x1, x2) van be angularly averaged' is a typo for 'can be'.","section":"Appendix E"},{"comment":"'according to it’s definition' should be 'its definition'.","section":"Appendix E"},{"comment":"'polysterene' should be 'polystyrene'.","section":"p. 3"},{"comment":"The manuscript uses a nonstandard accent in 'L´ evy'; the standard form is 'Lévy'.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The single-particle demonstration is convincing and within the journal's scope; the main obstacles are the missing turn-angle distribution, the typo in Eq. (D5), and the lack of quantitative uncertainties for the Lévy and collective exponents. I would be willing to see a revision that addresses these points."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Three things to know. First, the mechanism is genuinely new: pulsing the field lets a Quincke roller depolarize and repolarize, so the next rotation axis is (they claim) independent of the previous run direction, and the tumble duration is set by τT rather than by fluid relaxation. Second, the single-particle validation is unusually clean: the MSD and VACF are compared with published analytic expressions from Angelani and Detcheverry using measured V and τ and prescribed τT, with no fitting parameters. The agreement in Figs. 1b and 2c,d,h,i is visually strong and supports the exponential and power-law run-time programs. Third, the collective-dynamics part is exploratory and should be read as such.\n\nThe soft spot the stress-test flags is real. α = <cos Δθ> ≈ 0 is necessary but not sufficient for a uniformly random tumble angle. A symmetric distribution with equal probability of ±90° turns also gives α = 0, and because the MSD/VACF of a renewal walk depend only on the mean cosine, those fits cannot distinguish such a distribution from uniform. So the 'truly random reorientation' statement is not yet established. The fix is straightforward: show the measured distribution of Δθ, or at least its second moment, for τT/τmw ≫ 2. The physical argument about complete depolarization is plausible, and the sharp transition in α near τT/τmw ≈ 2 is consistent, but the direct evidence is missing. Relatedly, 'emulate any random walk' overstates the demonstrated scope; the paper shows two run-time distributions, not a general capability. The citation pattern is fine — the prior theory is properly attributed and no suspicious self-citation load.\n\nThe collective section is the weakest part. The exponents a ≈ 0.84–0.89 and the energy-spectrum slope −8/3 are presented without error bars or fit statistics, and the cluster-identification threshold (1.4d–1.6d) is an input that could affect the cluster statistics. These results are suggestive, and the qualitative resemblance to bacterial clusters is nice, but they are not quantified rigorously enough for the claims to stand as they are. This is a 'tighten the statistics' problem, not a fatal flaw.\n\nBottom line: this is a solid experimental paper with a real new capability. A serious referee should engage with it, and the revision should add the tumble-angle distribution, error bars on the collective exponents, and ideally data/code release. I'd take it to our reading group and would cite it once the distribution test is in; even as a preprint it's the clearest demonstration I know of a programmable active random walker.","headline":"A programmable Quincke-roller random walker with real experimental chops; the core single-particle result is solid, but the 'truly random tumble' claim needs a direct distribution test and the collective exponents need error bars.","tokens_in":16643,"tokens_out":3640,"would_cite":true,"duration_ms":32825,"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":"A pulsed electric field turns a rolling colloid into a programmable random walker.","keywords":["Quincke rotation","run-and-tumble motion","Lévy walk","active colloids","Maxwell-Wagner relaxation","mean squared displacement","collective dynamics","active matter"],"falsifier":"Measure the full distribution of turn angles $\\Delta\\theta$ for off-times with $\\tau_T/\\tau_{\\mathrm{mw}} \\gg 1$; if the histogram is not flat on the circle, or if the next run direction correlates with the previous run direction beyond the mean cosine, then the runs are not independent and the claim that any random walk can be emulated fails.","tokens_in":15702,"feed_emoji":"⚡","tokens_out":7978,"duration_ms":71315,"temperature":0.7,"pith_summary":"This paper reports a synthetic colloid whose trajectory can be programmed to match the random walks used by swimming bacteria. A micron-scale sphere rolls in a steady electric field through the Quincke effect; switching the field off and on again makes it stop, reorient, and start a new run. By choosing how long the field stays off relative to the particle's charge-relaxation time, the authors make consecutive runs either independent or correlated, and by drawing run durations from an exponential or power-law distribution they reproduce run-and-tumble motion and Lévy walks. Single-particle mean-squared displacements and velocity autocorrelations agree with the analytic predictions for those walks, and populations of the walkers form swarms, rotating clusters, and turbulent-like aggregates reminiscent of bacterial suspensions. The payoff is a tabletop system in which the microscopic motility rule, not just the particle density, can be tuned at will.","feed_headline":"Pulsed fields turn rolling colloids into tunable random walkers","feed_subtitle":"A pulsed field makes one colloid walk like a bacterium, putting bacterial-style swarming under direct lab control.","key_machinery":"The central mechanism is the Quincke instability combined with Maxwell-Wagner polarization relaxation. Quincke rotation is the spontaneous spinning of a polarized sphere around an axis perpendicular to the applied field; because that axis is degenerate in the plane perpendicular to the field, each re-polarization can pick a new direction. The paper uses the Maxwell-Wagner time $\\tau_{\\mathrm{mw}}$, the exponential time scale for induced surface charge to build up or decay, as the memory knob: off-times long compared with $\\tau_{\\mathrm{mw}}$ erase the previous orientation, while shorter off-times leave partial polarization that biases the next run. A programmable waveform generator turns this physics into a random-walk synthesizer by drawing $\\tau_R$ from a target distribution and setting pulse durations accordingly.","core_discovery":"The central claim is that the classical Quincke roller, a dielectric sphere that spins and rolls when polarized in a DC electric field, can be converted into a random walker whose every run and tumble is specified in advance by the applied voltage waveform. When the field is on, the sphere rolls straight at a speed set by the field amplitude; when the field is off, it stops and discharges on the Maxwell-Wagner time $\\tau_{\\mathrm{mw}}$, and when the field returns the Quincke instability selects a new rotation axis. If the off-time $\\tau_T$ is much larger than $\\tau_{\\mathrm{mw}}$, the new direction is stated to be fully randomized and the run and turn phases are independent; tuning $\\tau_T/\\tau_{\\mathrm{mw}}$ near or below 2 introduces a controlled directional memory. Drawing run durations $\\tau_R$ from a chosen probability distribution and encoding them as pulse widths yields run-and-tumble walks (exponential $\\tau_R$) and Lévy walks (power-law $\\tau_R$), with measured mean-squared displacement and velocity autocorrelation matching the analytic expressions for constant-speed walkers with finite turning time. The paper further reports that populations of these walkers reproduce collective signatures of bacterial suspensions, including anomalous number fluctuations and an energy spectrum scaling of $-8/3$.","pith_inferences":["Editorial inference: if the turn-angle distribution is confirmed to be uniform, the walker could serve as a programmable random-walk generator for testing optimal-search theories, for instance whether Lévy walks beat run-and-tumble in obstacle fields, without relying on live bacteria.","Editorial inference: the global field clock imposes synchronized runs and stops on all particles, a feature absent in bacterial suspensions; matching bacterial clustering statistics may therefore arise from a different mechanism than biological coordination, and comparing the two could separate clock-driven from interaction-driven ordering.","Editorial inference: scaling the same protocol to smaller colloids or lower speeds would introduce Brownian noise, yielding a controlled interpolation between the deterministic run-and-tumble regime and active Brownian motion."],"forward_implications":["One experiment can now generate ordinary random walks, run-and-tumble walks, and Lévy walks from the same colloid, with the effective diffusion coefficient set by field amplitude and pulse timing.","The run speed depends only on the field amplitude, so speed and walk statistics are independently tunable.","Populations of these walkers show collective phases seen in bacterial suspensions—swarms, rotating clusters, polar clusters, and disordered clusters—with number fluctuations more anomalous than equilibrium and an energy spectrum with $-8/3$ scaling.","Because every particle runs and stops on the same clock, the system provides a controlled experimental platform for testing theories that link single-particle motility patterns to emergent collective order.","The same waveform approach extends to alternating speeds and to waiting-time distributions that yield anomalous subdiffusion, and to other Quincke-powered particles such as helical propellers."],"supporting_citations":[{"why":"Establishes the Quincke rotation phenomenon that powers the straight runs.","marker":"[27]"},{"why":"Supplies the polarization relaxation equation and Maxwell-Wagner time that set the memory and depolarization scale.","marker":"[30]"},{"why":"Provides the analytic mean-squared displacement and velocity autocorrelation for run-and-tumble and Lévy walks against which the experiments are compared.","marker":"[32]"},{"why":"Provides the generalized run-and-turn formulation used to derive the MSD and VACF expressions for arbitrary run-time distributions.","marker":"[62]"},{"why":"Introduces Quincke rollers and their collective directed motion, the baseline that the random walker extends.","marker":"[24]"},{"why":"Documents emergent vortices and swarming in populations of Quincke rollers, compared with the collective phases reported here.","marker":"[25]"},{"why":"Gives the experimental system context, including surfactant tuning of the Maxwell-Wagner time.","marker":"[26]"},{"why":"Defines the Lévy-walk distribution with resting periods used to draw run times in the experiments.","marker":"[33]"}],"fun_headline_variants":["Pulsed fields craft colloid walkers with tailored runs and tumbles","Pulsed voltage programs colloid trajectories: run, tumble, repeat","Making colloids walk on command: from Lévy to run-and-tumble","Colloids with programmable random walks mimic bacterial swarming","Pulsed-field colloids: every run and tumble is set in advance"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that a fully depolarized colloid tumbles by picking its next direction uniformly at random; the reported evidence is only that the average cosine of the turning angle is near zero, which does not distinguish a uniform distribution from symmetric but non-uniform ones.","fun_headline_variants_meta":{"raw":{"variants":["Pulsed fields craft colloid walkers with tailored runs and tumbles","Pulsed voltage programs colloid trajectories: run, tumble, repeat","Making colloids walk on command: from Lévy to run-and-tumble","Colloids with programmable random walks mimic bacterial swarming","Pulsed-field colloids: every run and tumble is set in advance"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000778,"raw_usage":{"total_tokens":3445,"prompt_tokens":953,"completion_tokens":2492,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":569,"completion_tokens_details":{"reasoning_tokens":2398}},"tokens_in":569,"tokens_out":2492,"duration_ms":19086,"temperature":1.0,"reasoning_tokens":2398,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:50:35.150441+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the full distribution of turn angles $\\Delta\\theta$ for off-times with $\\tau_T/\\tau_{\\mathrm{mw}} \\gg 1$; if the histogram is not flat on the circle, or if the next run direction correlates with the previous run direction beyond the mean cosine, then the runs are not independent and the claim that any random walk can be emulated fails.","supporting_citations":[{"cited_title":"Angelani","cited_arxiv_id":null,"evidence_quote":"Provides the analytic mean-squared displacement and velocity autocorrelation for run-and-tumble and Lévy walks against which the experiments are compared."},{"cited_title":"Generalized run-and-turn mo- tions: From bacteria to L´ evy walks","cited_arxiv_id":null,"evidence_quote":"Provides the generalized run-and-turn formulation used to derive the MSD and VACF expressions for arbitrary run-time distributions."},{"cited_title":"Emergence of macroscopic directed motion in popula- tions of motile colloids","cited_arxiv_id":null,"evidence_quote":"Introduces Quincke rollers and their collective directed motion, the baseline that the random walker extends."},{"cited_title":"Emergent vortices in popula- tions of colloidal rollers","cited_arxiv_id":null,"evidence_quote":"Documents emergent vortices and swarming in populations of Quincke rollers, compared with the collective phases reported here."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the experimental system context, including surfactant tuning of the Maxwell-Wagner time."},{"cited_title":"L´ evy walks","cited_arxiv_id":null,"evidence_quote":"Defines the Lévy-walk distribution with resting periods used to draw run times in the experiments."}],"review_version":1}