{"id":"6a2bea23-e4f7-40f5-9b8b-3d7fe111d828","arxiv_id":"2501.14088","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"In bacteria-filled droplets, the effective bath diffusivity of passive tracers collapses onto one curve when plotted against bacterial density times the available space divided by tracer radius.","lead":"The authors placed tiny tracer beads inside water droplets filled with swimming bacteria and tracked their motion in 3D. They show the bacterial agitation acts like a noisy bath with a memory time of roughly 0.1 to 0.6 seconds, and that its diffusivity grows with bacterial density and the free space around the tracer.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The three-decade D_b collapse should be reproduced by u_b alone if tau_b is constant; that cross-plot is missing and is needed to rule out a fitting artifact.","rationale":"The reader's weakest assumption is the uniform, isotropic, single-timescale OUP model. I partially agree. The model assumption is central, but the more immediately testable and potentially damaging issue is internal consistency of the extracted parameters: the paper claims tau_b is roughly constant while u_b is highly scattered, yet D_b collapses over three decades. Since D_b = u_b^2 tau_b, a constant tau_b would force the collapse to be present in u_b alone. The absence of a u_b-versus-nR/R_i plot is a gap that can be closed without new experiments. If that plot shows the same collapse, the concern is resolved; if not, the conclusion that the bath diffusivity is controlled by nR/R_i would be an artifact of the fitting procedure or of tau_b variations disguised as 'roughly constant'. I keep the verdict CONDITIONAL because the requested check is feasible and would either strengthen the paper or require reinterpretation. I do not see grounds for rejection: the data are carefully acquired, the MSD fits look reasonable, and the authors acknowledge the lack of a theoretical derivation. The condition should be: provide the u_b cross-plot and/or a synthetic parameter-recovery test demonstrating that the three parameters are not degenerate.","tokens_in":12568,"tokens_out":7980,"duration_ms":73956,"concrete_test":"Plot the fitted bath speed u_b against nR/R_i for all data points, using the same symbols and axis treatment as Fig. 4. If u_b collapses onto a single curve over a comparable range, the D_b collapse is internally consistent and the tau_b-constant claim is supported. If the u_b plot is significantly more scattered than the D_b plot, then tau_b is not actually constant and the D_b collapse is an artifact of fitting correlations; the central scaling claim would then need revision or a full parameter-uncertainty analysis.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central empirical claim is that D_b = u_b^2 tau_b collapses over three decades when plotted against nR/R_i, with tau_b 'roughly constant' (0.1-0.6 s). If tau_b is indeed approximately constant, this collapse must already be present in u_b^2 (or u_b) plotted against the same variable. The paper, however, only shows u_b against n (Fig. 3b), where the data are highly scattered, and never shows u_b against nR/R_i. This omission is load-bearing: if u_b alone does not collapse on nR/R_i, then the D_b collapse is not a property of the bath velocity but is manufactured by the fitted tau_b values varying systematically (or by parameter trade-offs in the three-parameter fit), contradicting the claim that tau_b is constant. Conversely, if u_b does collapse, the paper should say so, since it would make the argument much stronger. A related concern is that no uncertainties or uniqueness checks are reported for the fitted parameters; the fit scans a grid of about 40,000 simulations and reports the global minimum of chi^2, but not the shape of the chi^2 landscape, so a degenerate ridge in (tau_b, u_b) space could produce an apparent collapse that mirrors the input variable nR/R_i.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript studies passive spherical tracers (solid beads or oil droplets) confined inside water droplets that contain swimming E. coli. From 2D and 3D particle tracking, the authors extract planar mean-square displacements and model the bacterial velocity field felt by the tracer as an isotropic Gaussian Ornstein-Uhlenbeck process with characteristic speed u_b and persistence time τ_b, while the spherical confinement is rendered as a reflective boundary. A numerical solution of this stochastic model is fitted to each experimental MSD using three free parameters (τ_b, u_b, and a sedimentation time τ_s), after a detection-noise correction. The reported results are that τ_b is roughly constant (0.1–0.6 s), u_b increases weakly with bacterial density, and the derived bath diffusivity D_b = u_b^2 τ_b collapses over about three decades when plotted against nR/R_i, where R is the available space between the tracer and the confining wall. The paper interprets this as evidence that a confined bacterial suspension acts as a colored-noise bath with a finite memory and a confinement-tunable intensity.","tokens_in":12840,"tokens_out":3693,"duration_ms":37962,"significance":"If the central empirical claim holds, the paper provides a useful quantitative characterization of confined active baths and a potentially general scaling relation involving bacterial density and geometric confinement. The work has several genuine strengths: it combines two experimental tracking platforms, includes a flat-chamber control system, explicitly corrects for detection noise rather than smoothing trajectories, verifies that trajectories reach the ergodic limit through the saturation plateau, and compares the fitted MSDs with a numerical model rather than relying on a harmonic approximation that the authors show is inapplicable to their data. These features make the study a solid experimental contribution to the active-bath literature. The main risk is in the inference chain from fitted parameters to the claimed three-decade collapse, because D_b is not measured directly but is constructed from fitted u_b and τ_b, and the manuscript does not yet demonstrate that this construction is robust and unique.","major_comments":[{"comment":"The central claim that D_b = u_b^2 τ_b collapses as a function of nR/R_i is not yet demonstrated convincingly, because D_b is a derived product of two fitted quantities and τ_b is only 'roughly constant' over a factor of about 6 (0.1 to 0.6 s). If τ_b is truly approximately independent of the control parameters, the same collapse should already be visible in u_b^2 (or u_b) plotted against nR/R_i. The manuscript does not show this plot, even though it does show u_b against n (Fig. 3b), where the data are highly scattered. Please report u_b and τ_b separately against nR/R_i, with confidence intervals, and quantify the residuals of the claimed power-law/collapse in D_b. Without this diagnostic, the collapse could in principle be manufactured by a systematic variation of the fitted τ_b or by parameter trade-offs in the three-parameter fit.","section":"Active bath diffusivity, Fig. 4"},{"comment":"No uncertainties, uniqueness checks, or identifiability tests are reported for the fitted parameters. The protocol scans about 40,000 simulations and selects the global minimum of χ^2, but the shape of the χ^2 landscape is not shown, so the reader cannot assess whether τ_b, u_b, and τ_s are independently constrained or whether a degenerate ridge exists. This is load-bearing because the main scaling law is built from fitted u_b and τ_b. Please provide confidence intervals (for example by bootstrapping over trajectories or over fit realizations), show the correlation structure among the fitted parameters, and demonstrate with synthetic MSD data generated from the model that the fitting procedure recovers the input parameters without systematic bias or degeneracy.","section":"Bath parameters extraction and Methods F"},{"comment":"The model assumes that the bacterial velocity felt by the tracer is spatially uniform, isotropic, Gaussian, and characterized by a single correlation time τ_b. This assumption is central to the definition of both τ_b and D_b, and the paper does not provide any direct validation of the Gaussian or single-timescale property, beyond the visual agreement of the fitted MSDs. I would like to see a concrete test: for example, a comparison of the experimental short-time velocity distribution (or an autocorrelation of the measured velocity increments) with the model prediction, or a statement of how strongly the extracted parameters and the nR/R_i scaling would change under an alternative noise model, such as one with two correlation times or non-Gaussian velocity statistics. This would clarify whether the extracted τ_b and D_b are intrinsic bath properties or merely effective outputs of the assumed model family.","section":"Numerical solution of a 3D stochastic model, Eq. (2)"}],"minor_comments":[{"comment":"Reference 18 appears to contain a typo: 'PNAS 1O7, 9541' should presumably read 'PNAS 107, 9541'.","section":"References"},{"comment":"The caption states that the color bar is common to panels (c) and (d), but the figure as described in the text does not make clear which quantity the color represents; please state it explicitly in the caption.","section":"Fig. 3"},{"comment":"The detection-noise correction assumes strictly delta-correlated noise in Eq. (12). It would be useful to state explicitly that neighboring-frame noise correlations were checked to be negligible, since the correction offset in Eq. (11) would otherwise have a different i-dependence.","section":"Materials and Methods E"},{"comment":"In the sentence 'The harmonic model ... underestimates the particle bath reduction', the wording is ambiguous; it appears to mean that the harmonic model overestimates the ratio u_p/u_b. Please clarify to avoid misreading.","section":"Particle velocities"},{"comment":"The conclusion states that τ_b shows no dependence on R_i or R_o, but the supporting evidence is only the text description in Fig. 3(a); a plot of τ_b against these control variables, or against nR/R_i, would make the statement directly verifiable.","section":"Conclusions"}],"recommendation":"major_revision","confidential_remarks":"The paper addresses a timely and interesting problem and the experimental effort is substantial. My recommendation of major revision is driven by the fact that the three-decade collapse, which is the central new claim, is presented without the standalone plot of u_b versus nR/R_i that would rule out a fitting artifact, and without any uncertainty or identifiability analysis for the three-parameter fit. These are fixable within the manuscript's scope, so I do not recommend rejection; but the current evidence is not yet sufficient to establish the headline scaling law."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The new thing is the three-decade collapse of Db = u_b^2 tau_b against nR/R_i for passive tracers in bacteria-filled droplets. If it holds, it gives a compact design rule for mixing in droplet bioreactors and extends earlier flat-chamber results to spherical confinement. The experiments are careful: multiple tracking setups, a 3D Lagrangian technique, explicit correction for detection noise, and a flat-chamber reference that sits sensibly on the same trend.\n\nThe paper is honest about its own limits: the OUP description is assumed, not derived, and the authors state the scaling expression 'still awaiting a complete theoretical explanation.'\n\nThe soft spot is the one the stress-test flags. Db is a product of two fitted parameters, u_b and tau_b. The paper claims tau_b is roughly constant (0.1-0.6 s), yet it never shows u_b (or u_b^2) against nR/R_i. If tau_b is truly roughly constant, the Db collapse should already be visible in u_b^2 alone; showing that plot would either make the argument much stronger or reveal that tau_b variations are doing real work. Without it, the collapse is under-supported. Relatedly, the fitting procedure scans ~40,000 simulations and reports the global chi^2 minimum, but not the shape of the chi^2 landscape, so a degenerate ridge in (tau_b, u_b) space could produce an apparent mirror of the input variable. No uncertainties are reported. These are fixable in revision: plot u_b vs nR/R_i, report bootstrap confidence intervals on the fitted parameters, and show at least one iso-chi^2 contour.\n\nNone of this kills the paper. The central claim is plausible and the data are not obviously over-fitted. The flat-chamber comparison provides an independent anchor, and the wide parameter range (n, R, R_i) is genuinely explored. The main result is a new experimental scaling law with clear practical relevance. I'd send this to a serious referee; the authors are clearly capable of addressing the concerns, and the community will want the cross-plot one way or the other.","headline":"Useful experimental scaling law for bacterial bath diffusivity in droplets, but the missing u_b cross-plot and lack of error bars leave the parameter extraction under-checked.","tokens_in":13392,"tokens_out":3154,"would_cite":true,"duration_ms":27572,"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":"The paper establishes that a confined bacterial suspension acts on a passive tracer as colored noise with a constant memory time of 0.1–0.6 s and a diffusivity that collapses three decades when plotted against $nR/R_i$.","keywords":["active bath","bacterial suspension","colored noise","Ornstein–Uhlenbeck process","enhanced diffusion","confinement","tracer diffusion","droplet"],"falsifier":"Measure the tracer's velocity autocorrelation directly from 3D tracks: if it does not decay as a single exponential with $\\tau_b$ between 0.1 and 0.6 s for all wall distances and densities, or if the velocity-increment distribution shows non-Gaussian tails, the Ornstein–Uhlenbeck bath description is falsified.","tokens_in":12395,"feed_emoji":"🦠","tokens_out":9116,"duration_ms":76243,"temperature":0.7,"pith_summary":"The paper asks what a passive particle feels when trapped inside a droplet of swimming bacteria, and answers that the bacterial suspension acts as a stochastic bath with temporal memory rather than a thermal bath. By tracking buoyant tracers (solid beads and oil droplets) over a wide range of droplet radii, tracer radii, and bacterial densities, the authors show that the tracer motion is reproduced by an Ornstein–Uhlenbeck noise with a correlation time $\\tau_b$ that stays roughly constant between 0.1 and 0.6 s. The bath diffusivity $D_b = u_b^2\\tau_b$ grows linearly with $nR/R_i$, where $n$ is bacterial density, $R=R_o-R_i$ is the available space, and $R_i$ is the tracer radius; data from all experiments collapse onto one curve over three decades. If this holds, then a droplet's ability to stir and transport an encapsulated object is set by a single confinement–density variable, which is directly relevant to droplet bioreactors, cloud-droplet chemistry, and cell-like crowded environments.","feed_headline":"Bacterial droplets act as colored-noise baths tunable by space","feed_subtitle":"Their diffusivity grows with density and available space, collapsing three decades of data onto one curve.","key_machinery":"The load-bearing object is the Ornstein–Uhlenbeck process (OUP), a Gaussian stochastic velocity field with correlation $\\langle u_i(t)u_j(t')\\rangle = u_b^2\\delta_{ij}e^{-|t-t'|/\\tau_b}$. The tracer follows $\\dot{\\mathbf r} = \\mathbf u - v_s\\hat z$ inside a reflective spherical boundary, and the model is solved numerically over roughly 40,000 parameter pairs $(\\tilde\\tau_b,\\tilde\\ell_b)$; the experimental planar MSD is matched to the simulated MSD by minimizing a log-scale chi-squared with the sedimentation time as a third fit parameter. This machinery converts complicated swimmer hydrodynamics into two effective bath parameters, $\\tau_b$ and $D_b=u_b^2\\tau_b$, whose values can then be compared across geometries.","core_discovery":"The central claim is that momentum transfer from a confined bacterial suspension to a passive tracer is described by colored noise with $\\langle u_i(t)u_j(t')\\rangle = \\frac{D_b}{\\tau_b}\\delta_{ij} e^{-|t-t'|/\\tau_b}$, where $\\tau_b$ is a finite memory time and $D_b = u_b^2\\tau_b$ is the bath diffusivity. Extracting these parameters from experimental planar mean-square displacements via a 3D stochastic model, the paper reports $\\tau_b$ between 0.1 and 0.6 s with no systematic dependence on bacterial density, tracer radius, or droplet radius, while $D_b$ increases with density and with the available space $R=R_o-R_i$. All measured $D_b$ values collapse over three decades when plotted against $nR/R_i$, indicating a linear dependence on bacterial concentration modulated by confinement. The paper therefore concludes that the active bath inside a droplet is characterized by a constant memory time and a confinement-tunable intensity.","pith_inferences":["An implication the paper leaves implicit: because $D_b$ grows with $R=R_o-R_i$, the same bacterial density produces a weaker bath in smaller droplets, so droplet-based bioreactors and emulsion mixers should exhibit a size-dependent stirring threshold that can be tested by measuring transport or reaction rates.","The constancy of $\\tau_b$ hints that the memory is set by single-swimmer encounter kinematics (swim speed, body length) rather than by collective motion; a test would be to alter swimmer speed chemically and see $\\tau_b$ change while $D_b$ keeps the $nR/R_i$ scaling.","The OUP is a coarse-grained effective description. Near the collective-motion density the linear law should saturate and the velocity increments should develop non-Gaussian tails; the flat-chamber $R\\to\\infty$ data already show where the collapse breaks down, so re-measuring at higher $nR/R_i$ would reveal the transition.","The combination $nR/R_i$ has the form of a density times available volume per tracer surface, suggesting a connection between $D_b$ and the mechanical swim pressure exerted on the confining interface; expressing $D_b$ as an active stress times a geometric factor would link bath diffusivity to droplet rheology."],"forward_implications":["All confined-bath experiments collapse onto a single master curve $D_b \\propto nR/R_i$, so one scaling relation organizes transport measurements across different droplet and tracer sizes.","Confinement is a tuning parameter: at fixed bacterial density, reducing the available space $R=R_o-R_i$ weakens the bath's diffusivity, so droplet size alone controls mixing intensity.","The bath memory time $\\tau_b\\approx 0.1$–$0.6$ s is independent of bacterial density, tracer size, and droplet size, meaning the short-time persistence of the active forcing is an intrinsic property of the swimmer–tracer interaction.","The effective friction coefficient extracted from the fits exceeds Stokes drag and grows with the confinement ratio $R_i/R_o$, independently of bacterial density, indicating wall lubrication rather than activity sets the tracer's resistance."],"supporting_citations":[{"why":"Establishes the founding observation that motile bacteria enhance passive-tracer diffusion beyond thermal Stokes–Einstein values, the effect this paper extends to spherical confinement.","marker":"(2)"},{"why":"Supplies the Ornstein–Uhlenbeck active-bath model and its harmonic-trap MSD solution, the starting point the authors generalize with a full 3D spherical model.","marker":"(28)"},{"why":"Describes the in-house 3D Lagrangian tracking technique that provides the full (x,y,z) bead trajectories used to characterize the bath.","marker":"(40)"},{"why":"Provides earlier measurements of tracer kinematics in bacterial suspensions whose time scales are compared with the extracted $\\tau_b$.","marker":"(11)"},{"why":"Shows that a harmonic-potential approximation fails beyond small normalized variances, motivating the numerical 3D stochastic model used to extract bath parameters.","marker":"(45)"},{"why":"Documents tracer-size dependence in active-fluid diffusion, supporting the inclusion of $R_i$ in the confinement variable $nR/R_i$.","marker":"(50)"},{"why":"Measures droplets driven by enclosed bacteria, supplying the droplet-confinement hydrodynamics and comparison for bath parameters.","marker":"(51)"},{"why":"Provides the reflective-boundary simulation scheme used to confine the particle trajectory to the spherical droplet.","marker":"(46)"}],"fun_headline_variants":["Bacterial droplet baths: diffusivity scales with space and density","Colored noise from bacteria in droplets, tunable by confinement","Bacterial baths in droplets: memory constant, intensity flexible","Confined bacteria give tracers a boost that scales with room","Drop bacteria: bath memory fixed, diffusion scales with space"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the tracer feels a single uniform, Gaussian, exponentially correlated random push, so one correlation time and one noise strength describe the whole bath; if the bacterial forcing varies with position, depends on the wall, or has several timescales, the extracted $\\tau_b$ and $D_b$ are effective fit parameters, not intrinsic properties.","fun_headline_variants_meta":{"raw":{"variants":["Bacterial droplet baths: diffusivity scales with space and density","Colored noise from bacteria in droplets, tunable by confinement","Bacterial baths in droplets: memory constant, intensity flexible","Confined bacteria give tracers a boost that scales with room","Drop bacteria: bath memory fixed, diffusion scales with space"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000548,"raw_usage":{"total_tokens":2626,"prompt_tokens":963,"completion_tokens":1663,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":579,"completion_tokens_details":{"reasoning_tokens":1579}},"tokens_in":579,"tokens_out":1663,"duration_ms":12279,"temperature":1.0,"reasoning_tokens":1579,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T15:23:15.773076+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the tracer's velocity autocorrelation directly from 3D tracks: if it does not decay as a single exponential with $\\tau_b$ between 0.1 and 0.6 s for all wall distances and densities, or if the velocity-increment distribution shows non-Gaussian tails, the Ornstein–Uhlenbeck bath description is falsified.","supporting_citations":[],"review_version":1}