{"id":"ac361d28-3dd7-4919-97a4-d4443cb3bbca","arxiv_id":"2505.15396","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Finite, material-intrinsic sensory delay in photoconductive Janus colloids enhances their localization in dark regions of spatiotemporal light patterns, with an optimal delay growing linearly with pattern size.","lead":"Photoconductive Janus colloids driven by AC electric fields slow down over about 2 seconds after UV light is removed, while they speed up almost instantly when light returns. This asymmetric sensory delay lets the particles pile up more strongly in dark regions of projected light patterns than the standard instantaneous-response rule predicts, offering a new handle for patterning active matter.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Simulation model uses a single symmetric switching rate for velocity adaptation, but the measured response is strongly asymmetric (deceleration ~2 s, acceleration near-instant); the quantitative design rules in Fig. 5 may therefore not apply to the experimental system.","rationale":"The reader's weakest assumption identifies the same load-bearing concern: the simulation's switching kinetics are under-specified relative to the measured asymmetric response. This is the right point of attack because the paper's headline quantitative result—that sensory delay enhances localization and has an optimal value v_H τ* ≈ 0.41L—rests on simulations that use a single rate 1/τ for transitions in both directions. Since the real system only has a substantial delay when slowing down, a symmetric model would add an artificial slow-acceleration process that would increase time spent in bright regions and reduce the localization ratio, potentially shifting the optimal delay and the fitted slope. The experimental data in Fig. 2 provide strong, independent support for the existence and qualitative role of the delay, and the over-localization trend in Fig. 4 is suggestive, but the absence of error bars and the unspecified model kinetics make the quantitative agreement hard to evaluate. The proposed rerun with asymmetric rates would directly settle whether the design rules survive the correct kinetics. I therefore see no need to change the reader's CONDITIONAL verdict; the concern is substantive but testable, not fatal.","tokens_in":14892,"tokens_out":9425,"duration_ms":90651,"concrete_test":"Rerun the Section IV D simulations with two distinct switching rates extracted from Fig. 2b: fast-to-slow rate 1/τ_off ≈ 0.5 s⁻¹ and slow-to-fast rate 1/τ_on at least 3 s⁻¹ (or effectively instantaneous), for the same v_L, v_H, L, and D_R values. Compare the resulting ρ_L/ρ_H versus τ curves, the optimal-delay scaling v_H τ* versus L, and the (ρ_L/ρ_H)_max and τ_1/2 maps with Figs. 4 and 5. If the asymmetric model reproduces the experiments and the 0.41L slope, the single-rate simplification is benign; if it does not, the design rules must be re-derived for the measured one-sided delay.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central quantitative claims—over-localization beyond ρ_L v_L = ρ_H v_H, the optimal-delay scaling v_H τ* ≈ 0.41 L, and the design maps in Fig. 5c,d—are produced by the numerical model described in Sections II C and IV D. There, each particle is said to switch between v_L and v_H 'at a rate proportional to 1/τ', with a single characteristic response time τ. But the measured response in Fig. 2b is strongly asymmetric: deceleration from high to low activity has τ ≈ 2 s, while acceleration from low to high is essentially instantaneous (τ below the 300 ms exposure time). If the simulation uses one symmetric rate, then particles entering illuminated regions from dark regions remain slow for an average time τ, increasing the residence time in high-activity regions and suppressing the density ratio ρ_L/ρ_H. The experimentally observed enhancement relies specifically on the asymmetry: particles penetrate dark regions while still fast, but do not linger in bright regions. A model that delays both transitions would produce systematically different ρ_L/ρ_H values and different optimal delays than the one-sided sensory delay actually realized. The Methods do not state whether two different rates are used, and the localization data in Fig. 4 lack error bars, so the apparent agreement between experiment and simulation cannot currently be quantitatively assessed. The qualitative phenomenon is well supported by Fig. 2 and the trajectory data, but the design rules and the fitted slope 0.39 in Fig. 5b are load-bearing on the unverified symmetry of the switching kinetics.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports an experimental system of silica-titania Janus microswimmers driven by induced-charge electrophoresis in AC electric fields, with propulsion speed controlled orthogonally by UV illumination through the photoconductivity of the titania cap. The authors characterize the velocity response to illumination steps and find a strongly asymmetric sensory delay: deceleration from high to low velocity takes about 2 s, while acceleration is nearly instantaneous. In checkerboard illumination patterns, they observe enhanced localization of particles in low-activity regions beyond the standard steady-state relation ρ_L v_L = ρ_H v_H, and they attribute this enhancement to the finite sensory delay. Using Brownian dynamics simulations with parameters extracted from experiments, they reproduce the qualitative trajectories and the trend in localization, and they derive a scaling law for the optimal delay, v_H τ* ≈ 0.41 L, from a geometric argument. They also provide design maps for maximum localization contrast and for the delay tolerance as functions of pattern size and velocity ratio. The central conceptual claim is that material-intrinsic sensory delay, unlike the effectively instantaneous response assumed in most synthetic active colloids, is a useful and tunable control parameter for density patterning.","tokens_in":15131,"tokens_out":3546,"duration_ms":32645,"significance":"If the quantitative results hold, this is a significant advance in active-matter control: it demonstrates a synthetic colloid whose propulsion speed is regulated by a material property (photoconductivity) orthogonal to the driving field, and it identifies sensory delay as a practical knob for density patterning, with explicit design rules. The experimental characterization of the asymmetric delay is clean and well documented, and the simulations use measured velocities and delay as inputs rather than fitting the localization data, so the over-localization is an emergent model prediction. The geometric scaling argument is transparent and matches the simulation slope closely. The paper also provides falsifiable design maps (Fig. 5c,d) that could guide material selection for other swimmers. However, the quantitative claims rest on a simulation model whose switching kinetics are not aligned with the measured asymmetric response, and the experimental localization data lack error bars; these weaknesses currently prevent a full quantitative assessment of the central claim of enhanced localization and of the specific design rules.","major_comments":[{"comment":"The simulation model is described as switching between the low and high velocities 'at a rate proportional to 1/τ', i.e., with a single characteristic response time. The measured response in Fig. 2b is strongly asymmetric: deceleration from high to low activity is exponential with τ ≈ 2 s, while acceleration from low to high activity is essentially instantaneous (below the 300 ms exposure time). A symmetric two-state model delays both transitions, so particles entering an illuminated region remain slow for an average time τ, which reduces the density ratio ρ_L/ρ_H and shifts the optimal delay relative to a model with one-sided delay. The Methods do not state whether different rates are used for the two transitions, and the quantitative results in Figs. 4 and 5 are produced by this model. The authors should implement the measured asymmetric kinetics in the simulations and show whether the localization ratios and the fitted slope in Fig. 5b are preserved, or justify explicitly why a symmetric model captures the relevant physics despite the measured asymmetry.","section":"Section IV D, Eq. (4)-(5), Fig. 2b"},{"comment":"The experimental localization ratios ρ_L/ρ_H in Fig. 4 are shown as single square markers with no error bars, no confidence intervals, and no stated number of particles or frames used for the steady-state average. Because the central quantitative claim is that experiments exceed the instantaneous-response prediction ρ_L v_L = ρ_H v_H for larger patterns, the absence of statistical information makes it impossible to assess the strength of the agreement with the simulations, which is a load-bearing element of the paper. The authors should provide error bars (e.g., standard error over independent measurements or temporal block averages), state the number of particles tracked and the equilibration time, and clarify how the steady state was determined.","section":"Fig. 4"}],"minor_comments":[{"comment":"The geometric scaling argument is presented as a justification for the fitted slope 0.39 in Fig. 5b, but the derivation assumes purely ballistic motion and neglects rotational diffusion and translational noise, while the simulations include these effects. The paper does state that τ < 1/D_R in the investigated regime, which supports the ballistic approximation, but it would be helpful to state this explicitly in the derivation and to note that the geometric estimate is an approximation, so the close agreement with the fit is not by construction.","section":"Section II C, Fig. 5b"},{"comment":"Equation (7) in the supplementary information appears to contain a typographical error: the integrand is written as sqrt((x−y)^2 + L + 2 sqrt(x^2 + y^2)), whereas the intended expression is presumably sqrt((x−y)^2 + L^2) + 2 sqrt(x^2 + y^2). Please correct this.","section":"Supplementary S3, Eq. (7)"},{"comment":"In the caption of Fig. 5b, the phrase 'the hatched area are shows where' should be corrected to 'the hatched area shows where'.","section":"Fig. 5 caption"},{"comment":"The threshold pattern size for which finite delay enhances localization is defined through the persistence length in the slow region, v_L τ_R, with τ_R = 1/D_R. Since the paper earlier defines the persistence length as L_p = v/D_R, the notation τ_R could be confused with response time; consider writing it as v_L/D_R for clarity.","section":"Section II C"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"What you should know: this is the first synthetic colloidal system where a sensory delay in velocity adaptation is intrinsic to the material (photoconductivity of titania) and controlled orthogonally to the propulsion field (AC ICEP). The authors measure a ~2 s slow-down timescale and near-instant speed-up, and show in checkerboard patterns that this asymmetry produces density ratios beyond the instantaneous ρ_L v_L = ρ_H v_H benchmark. That over-localization is a real experimental effect, and the simulations reproduce it using measured velocities and delay as inputs rather than fitting to the localization data. The geometric scaling argument for the optimal delay, v_H τ* ≈ 0.41 L, is plausible and matches the simulation slope (0.39). Credit is due: the experiments are clean, the delay characterization is systematic, and the authors correctly link their result to earlier work on photokinetic bacteria and sensorial-delay robots.\n\nThe main soft spot is the simulation model's switching kinetics. The Methods say particles switch between v_L and v_H 'at a rate proportional to 1/τ' without stating whether the deceleration and acceleration transitions use different rates. Given that the measured response is strongly asymmetric (deceleration ~2 s, acceleration < 300 ms), a single symmetric τ would bias the residence times in bright regions and could change both the localization ratios and the optimal delay scaling. This is a documentation gap rather than a demonstrated error, but it is load-bearing for the design maps in Fig. 5c,d. Also, the localization data in Fig. 4 have no error bars, so the experiment-simulation agreement cannot be quantitatively assessed. Both issues are fixable with clarification and added error analysis.\n\nWho this is for: experimentalists in active matter who want a material-based knob for density patterning, and theorists who model adaptive microswimmers. It deserves a serious referee. I would send it out, with the expectation that the simulation details be made explicit.","headline":"Clean experimental demonstration of delay-enhanced localization in photoconductive Janus colloids; quantitative design rules need simulation clarification.","tokens_in":15737,"tokens_out":2562,"would_cite":true,"duration_ms":23343,"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 finite sensory delay in photoconductive Janus particles lets them over-accumulate in dark regions of a light pattern, beyond the standard instantaneous-response limit.","keywords":["Janus particles","induced-charge electrophoresis","photoconductivity","sensory delay","density localization","active colloids","motility control","light patterning"],"falsifier":"Measure $\\rho_L/\\rho_H$ versus pattern size across $L = 20$–$100\\,\\mu\\mathrm{m}$ for at least two deliberately different delay times (for example, caps with different charge-carrier lifetimes) and check whether the peak position follows $v_H \\tau^* \\approx 0.41L$; if the peak does not shift with $\\tau$ and $L$ in this way, the over-localization is not governed by the claimed ballistic-penetration balance.","tokens_in":14665,"feed_emoji":"🔬","tokens_out":8675,"duration_ms":73262,"temperature":0.7,"pith_summary":"Silica–titania Janus particles that swim by induced-charge electrophoresis in an AC field adjust their speed when UV light changes the conductivity of their titania caps. The paper's central claim is that the finite time this adjustment takes—about two seconds for slowing down—is a control handle: in checkerboard light patterns with feature sizes of 40 µm and larger, it produces steady-state densities in the dark regions that exceed the standard prediction $\\rho_L v_L = \\rho_H v_H$. The claim is supported by particle trajectories, density counts, and a Langevin simulation in which particles switch between two speeds at a rate $1/\\tau$. The result is a design rule: the best delay satisfies $v_H \\tau^* \\approx 0.41L$, with the proportionality set by the average ballistic crossing length of the pattern. If correct, material-intrinsic sensory delay gives synthetic swimmers a biological-style adaptive response without external feedback.","feed_headline":"Delayed braking lets active colloids pack dark regions","feed_subtitle":"A built-in two-second lag in titania-coated swimmers beats the standard instant-response density limit.","key_machinery":"The load-bearing object is the photoconductive titania cap on a Janus particle, whose UV-regulated conductivity changes propulsion speed independently of the AC field that drives ICEP; the sensory delay $\\tau$ is the characteristic time of that speed change. The paper treats the delay with a two-state velocity switch ($v_L \\leftrightarrow v_H$ at a rate proportional to $1/\\tau$), anchored to the measured exponential relaxation, and a geometric identity: the average ballistic path across a square of side $L$ is $L_{\\mathrm{eff}} \\approx 0.82L$, so optimal penetration is $v_H \\tau^* \\approx 0.5 L_{\\mathrm{eff}} \\approx 0.41L$. The baseline it beats is the exact non-interacting result $\\rho_L v_L = \\rho_H v_H$, which assumes instantaneous response. A measured asymmetry—slowing takes about 2 s, acceleration is nearly instantaneous—is reported, while the simulations use a single relaxation rate from Section IV D.","core_discovery":"The discovery is that a photoconductive cap with a seconds-long response time changes where active colloids accumulate. Measured against the steady-state rule $\\rho_L v_L = \\rho_H v_H$, which holds for instantaneously adapting swimmers, the particles over-localize in the low-activity (dark) squares of a checkerboard once the pattern size $L$ reaches about 40 µm. The localization ratio $\\rho_L/\\rho_H$ rises with the sensory delay $\\tau$, passes through a maximum, and then falls again; the optimal delay scales with pattern size as $v_H \\tau^* \\approx 0.41(L - L_{\\tau^*=0})$, matching a geometric estimate based on the average trajectory length $L_{\\mathrm{eff}} \\approx 0.82L$ through a square. The paper argues the mechanism is ballistic penetration: a fast particle crossing into a dark region keeps its high speed for a time $\\tau$, travels deeper than an instantly adapting swimmer would, and therefore spends longer escaping at the low speed. This turns the charge-carrier lifetime of the titania cap into a physical parameter for density patterning.","pith_inferences":["The measured asymmetry (fast acceleration, ~2 s deceleration) suggests that reversing the polarity of a moving pattern, or modulating uniform light in time, should produce direction-dependent density responses; the paper's symmetric two-state model likely underestimates such effects.","The same material-intrinsic delay idea should transfer to other stimuli with slow material responses, such as pH, temperature, or chemical concentration, where the relaxation time of the transducer replaces external feedback.","A direct test of the ballistic-penetration mechanism would be to compare localization for square versus stripe patterns of equal area, since the average crossing length differs and the optimal delay should shift accordingly.","If charge-carrier lifetime can be shortened or lengthened independently (for example, by hole scavengers or surface treatments), the design rule maps directly onto a material-selection procedure, effectively making sensory delay a tunable material property."],"forward_implications":["For pattern sizes $L \\geq 40\\,\\mu\\mathrm{m}$, choosing a material with the right delay gives a density contrast in dark regions beyond what the instantaneous inverse-velocity law allows.","The measured relation $v_H \\tau^* \\approx 0.41L$ gives a concrete target: engineers can tune the cap's charge-carrier lifetime to match a desired feature size.","For small patterns ($L \\approx 20\\,\\mu\\mathrm{m}$) or very short delays, the familiar $\\rho \\propto 1/v$ behavior is recovered, so the new effect appears only above a threshold set by the persistence length in the slow region.","Because speed control is orthogonal to propulsion, the same swimming mechanism can be used at different speeds without changing the electric field, simplifying pattern design.","The mapping of $(\\rho_L/\\rho_H)_{\\mathrm{max}}$ and $\\tau_{1/2}$ over $v_H$ and $L$ yields practical contrast-versus-resolution trade-offs for writing colloidal patterns."],"supporting_citations":[{"why":"Gives the exact steady-state result $\\rho(r) \\propto 1/v(r)$ that the paper takes as the instantaneous-response baseline.","marker":"[16]"},{"why":"Provides the light-induced active rectification theory that the checkerboard localization experiments extend.","marker":"[17]"},{"why":"Supplies the run-and-tumble statistical mechanics behind the density-velocity relation.","marker":"[18]"},{"why":"Demonstrates photoconducting Janus particles with optically modulated propulsion, the direct design basis for the present particles.","marker":"[20]"},{"why":"Previous feedback-controlled colloids with space-dependent dynamics whose simulation approach is adapted here to velocity control.","marker":"[22]"},{"why":"Shows dynamic density shaping of photokinetic bacteria, the biological analogue for patterning by motility control.","marker":"[24]"},{"why":"Explains long-lived trapped charge carriers in titania via adsorbed water, which underlies the seconds-long sensory delay.","marker":"[52]"},{"why":"Supplies the Langevin equations with memory used for the simulated trajectories in Figures 3-5.","marker":"[55]"}],"fun_headline_variants":["Delayed braking overpacks active colloids in dark","Sensory delay sharpens colloid dark-square localization","Lag time tunes where light-driven swimmers gather","Light-sensitive swimmers pile up with built-in delay"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The quantitative design rules from the simulations hold only if a single relaxation rate describes both speeding up and slowing down; in the measured response, slowing takes about two seconds while acceleration is nearly instantaneous, so a two-rate model could shift the predicted optimal delays and density ratios.","fun_headline_variants_meta":{"raw":{"variants":["Delayed braking overpacks active colloids in dark","Sensory delay sharpens colloid dark-square localization","Lag time tunes where light-driven swimmers gather","Light-sensitive swimmers pile up with built-in delay"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000301,"raw_usage":{"total_tokens":1753,"prompt_tokens":980,"completion_tokens":773,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":596,"completion_tokens_details":{"reasoning_tokens":712}},"tokens_in":596,"tokens_out":773,"duration_ms":7701,"temperature":1.0,"reasoning_tokens":712,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T15:18:28.301568+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure $\\rho_L/\\rho_H$ versus pattern size across $L = 20$–$100\\,\\mu\\mathrm{m}$ for at least two deliberately different delay times (for example, caps with different charge-carrier lifetimes) and check whether the peak position follows $v_H \\tau^* \\approx 0.41L$; if the peak does not shift with $\\tau$ and $L$ in this way, the over-localization is not governed by the claimed ballistic-penetration balance.","supporting_citations":[{"cited_title":"Frangipane, D","cited_arxiv_id":null,"evidence_quote":"Shows dynamic density shaping of photokinetic bacteria, the biological analogue for patterning by motility control."},{"cited_title":"Zehavi, D","cited_arxiv_id":null,"evidence_quote":"Demonstrates photoconducting Janus particles with optically modulated propulsion, the direct design basis for the present particles."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Previous feedback-controlled colloids with space-dependent dynamics whose simulation approach is adapted here to velocity control."},{"cited_title":"Litke, Y","cited_arxiv_id":null,"evidence_quote":"Explains long-lived trapped charge carriers in titania via adsorbed water, which underlies the seconds-long sensory delay."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the Langevin equations with memory used for the simulated trajectories in Figures 3-5."}],"review_version":1}