{"id":"a667f267-6840-4edd-bbcf-d65b5bed30c6","arxiv_id":"2607.05173","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":3,"one_line_summary":"An age-structured formulation restores Markovianity for semi-Markovian switching in a fluctuating harmonic trap, yielding exact steady-state moments and showing that injected power becomes universal in the stochastic-resetting limit while spatial fluctuations retain memory of the waiting-time tails.","lead":"The paper builds an exact theory for a Brownian particle in a harmonic trap whose stiffness switches at random times drawn from any distribution, not just exponential ones. It shows that while the energy cost of maintaining this system can become universal, the spatial spread of the particle remains highly sensitive to the specific timing statistics.","discovery_kind":"unclear","skeptic_critique":{"model":"glm-5.2","headline":"Eq. (45) contains a dimensional inconsistency: kBT/(K + 2Dτ_off) adds quantities with incompatible units. The correct birth moments are ⟨x²⟩b,off = kBT/K + 2Dτ_off(1/ηon − 1) and ⟨x²⟩b,on = kBT/K + 2Dτ_off/ηon. The universal injected power (Eq. 46) is nonetheless correct, verified by independent re-","rationale":"The reader identified normalizability of stationary densities for heavy-tailed waiting times as the weakest assumption. I find this concern does not actually land: for power-law r_off(a) ~ a^{-k} with k>2 (finite mean, as required by Eq. 2), the spatial distributions have tails n_off(x) ~ |x|^{3-2k} and n_on(x) ~ |x|^{1-2k} (Eqs. 64, 66), both normalizable since the exponents are < −1 for k > 2. Divergence of ⟨x²⟩ for 2 < k ≤ 3 does not imply non-normalizability of the density. The formalism only requires finite mean residence times (Eq. 2), which is explicitly assumed. The real issue I identified is the dimensional error in Eq. (45): kBT/(K + 2Dτ_off) is dimensionally inconsistent, and the correct expressions are ⟨x²⟩b,off = kBT/K + 2Dτ_off(1/ηon − 1) and ⟨x²⟩b,on = kBT/K + 2Dτ_off/ηon. This is likely a typographical error (misplaced fraction bar), but it breaks the derivation chain from Eq. (22) to Eq. (46). Despite this, the central claim (Eq. 46) is correct, as I verified by independent derivation from the general moment equations. The general framework is sound: Eq. (38) is verified, the virial result PE = kBT/2 (Eq. 30) follows from exact cancellation, and the SR limit follows correctly. The variance-space formulation (Sec. V) is internally consistent, with exact solutions for exponential and deterministic cases that can be checked directly. No code is provided for the numerical simulations (Figs. 2-4), but the analytical predictions are independently verifiable. The verdict remains CONDITIONAL: the central results are correct and novel, but the error in Eq. (45) should be corrected, and independent numerical verification would strengthen the work. The concern is presentation-level, not a fundamental flaw in the framework.","tokens_in":16517,"tokens_out":20546,"duration_ms":343736,"concrete_test":"Independently re-derive the SR-limit birth moments by substituting the K1→0 expansion of A/η1 (Eq. 22) and carefully evaluating the indeterminate form τK1·η1 → 2τ_off. Confirm that ⟨x²⟩b,off − ⟨x²⟩b,on = −2Dτ_off, which yields Eq. (46) via Eq. (37). Additionally, verify that the corrected Eq. (45) [kBT/K + 2Dτ_off(1/ηon − 1)] reduces to Eq. (48) for exponential waiting times (ηon = τon/(τon + τK/2)).","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim—universality of injected power in the SR limit (Eq. 46)—is correct. I verified it by independently re-deriving the birth moments from Eqs. (19)–(22) and taking K1→0 carefully. The two coupled moment equations are ⟨x²⟩b1 = DτK2·η2 + (1−η2)⟨x²⟩b2 and ⟨x²⟩b2 = DτK1·η1 + (1−η1)⟨x²⟩b1. In the SR limit (K1→0), τK1·η1 → 2τ_off (since η1 ≈ 2μK1τ1), yielding ⟨x²⟩b,off = DτK + 2Dτ_off(1/ηon − 1) and ⟨x²⟩b,on = DτK + 2Dτ_off/ηon. Their difference is −2Dτ_off, which via Eq. (37) gives ⟨Ẇ⟩ = DKτ_off/(τ_off+τ_on). This matches Eq. (46). However, Eq. (45) as written states ⟨x²⟩b,off = kBT/(K + 2Dτ_off)·(1/ηon − 1), which is dimensionally inconsistent: K has units [energy/length²] while Dτ_off has units [length²]. The correct expression is kBT/K + 2Dτ_off(1/ηon − 1), i.e., a misplaced fraction bar. As written, Eq. (45) does not reduce to the correct exponential limit (Eq. 48) and would not yield Eq. (46) if substituted into Eq. (37). The error is likely typographical, but it breaks the presented derivation chain from Eq. (22) to Eq. (46). The reader's normalizability concern is less pressing: for power-law waiting times with k>2, the spatial tails decay as |x|^{3−2k} (Eq. 64), ensuring normalizability even when ⟨x²⟩ diverges.","agreement_with_reader":"partial"},"referee_report":{"model":"glm-5.2","summary":"This manuscript develops an age-structured formulation for a Brownian particle in a harmonic trap whose stiffness switches between two values with arbitrary (semi-Markovian) waiting-time statistics. By augmenting the state space with an age variable, the authors restore Markovianity and derive a local Fokker–Planck description. From this, they obtain exact steady-state integral equations for spatial and birth distributions, exact second moments, and expressions for potential energy and injected power. A key result is that the average potential energy is universally PE = kBT/2 via a generalized virial relation, independent of switching statistics. In the stochastic-resetting (SR) limit (K1→0), the injected power becomes universal, ⟨Ẇ⟩ = DKτ_off/(τ_off + τ_on), depending only on mean residence times, while spatial fluctuations retain sensitivity to the full waiting-time distributions. A variance-space reformulation yields exact solutions for exponential and deterministic switching, asymptotic tails for algebraic waiting times, and exact moment hierarchies.","tokens_in":16930,"tokens_out":5281,"duration_ms":249598,"significance":"The paper makes several valuable contributions. The age-structured framework provides a clean, first-principles alternative to memory-kernel formulations for semi-Markovian switching, and the derivation is parameter-free (waiting-time distributions are inputs, not fitted quantities). The generalized virial relation (Eq. 31) giving PE = kBT/2 for arbitrary homogeneous potentials is an elegant and broadly applicable result. The universality of injected power in the SR limit, contrasted with the non-universality of spatial fluctuations, is a sharp and falsifiable prediction. The variance-space formulation is a creative reformulation that yields exact solutions and asymptotic results. Numerical simulations confirm the analytical predictions for variance and spatial distributions. The limiting-case checks (equilibrium K1=K2, Markovian exponential switching, fast-switching effective medium) are thorough.","major_comments":[{"comment":"Section IV, Eq. (45): The birth moments in the SR limit are written as ⟨x²⟩_{b,off} = kBT/(K + 2Dτ_off)·(1/η_on − 1) and ⟨x²⟩_{b,on} = kBT/(K + 2Dτ_off)·(1/η_on). This expression is dimensionally inconsistent: K has units [energy/length²] while Dτ_off has units [length²], so they cannot be added in a single denominator. Independent re-derivation from Eqs. (22)–(25) by taking K1→0 carefully yields the correct expressions ⟨x²⟩_{b,off} = kBT/K + 2Dτ_off(1/η_on − 1) and ⟨x²⟩_{b,on} = kBT/K + 2Dτ_off/η_on. The error appears to be a misplaced fraction bar. As written, Eq. (45) does not reduce to the correct exponential limit (Eq. 48) and does not yield Eq. (46) when substituted into Eq. (37). The universal result ⟨Ẇ⟩ = DKτ_off/(τ_off + τ_on) (Eq. 46) is nonetheless correct, as verified by the corrected expressions: the difference ⟨x²⟩_{b,off} − ⟨x²⟩_{b,on} = −2Dτ_off, which via Eq. (37) gives ","section":null},{"comment":"Section IV, Eq. (47): The expression ⟨x²⟩_{off} = ⟨x²⟩_{b,off} + (D/τ_off)⟨a²⟩_{off} depends on ⟨x²⟩_{b,off} from Eq. (45). Since Eq. (45) is incorrect as written, the consistency of Eq. (47) with Eq. (48) (the exponential limit) cannot be verified without first correcting Eq. (45). The authors should explicitly show that the corrected Eq. (45), when combined with Eq. (47), recovers Eq. (48) for exponential waiting times, thereby closing the derivation chain.","section":null}],"minor_comments":[{"comment":"Section III.C, Fig. 1: The caption states parameters τ_K1 = τ_K2 = τ_1 = τ_2 = 1, but the figure shows A_max = tanh(1) ≈ 0.7616. It would help to briefly state how A_max = tanh(1) follows from these parameter choices (it arises from the deterministic limit of the Gamma distribution with these parameters), so the reader can verify this value.","section":null},{"comment":"Section V, Eq. (53): The notation σ_t ∈ {0,1} for the switching variable is introduced without explicit definition of which value corresponds to which state. A brief clarification (σ_t = 0 for released, σ_t = 1 for confined, or vice versa) would improve readability.","section":null},{"comment":"Section V.B, Eq. (62): The uniform distribution p_off(s) = c_off is stated to hold on a specific interval, but the bounds involve c_off and c_on in a way that could be clearer. A brief sentence explaining the physical origin of these bounds (the minimum and maximum values of s reached during a deterministic cycle) would help.","section":null},{"comment":"The paper would benefit from a brief remark on the conditions under which the integral equations (15)–(16) admit normalizable solutions. The asymptotic analysis in Section V.B (Eq. 64) shows that for power-law waiting times with k > 2, the spatial tails decay as |x|^{3−2k}, ensuring normalizability even when ⟨x²⟩ diverges. Stating this explicitly near Eqs. (15)–(16) would address a natural concern about the well-posedness of the steady-state problem for heavy-tailed distributions.","section":null},{"comment":"Reference [25] appears to have an incomplete author list formatting ('L. B. P. P.-A. H. G. M. Rémi Goerlich, Minghao Li and C. Genet'). Please verify the author names.","section":null}],"recommendation":"major_revision","confidential_remarks":"The dimensional error in Eq. (45) is the primary concern. It is almost certainly a typographical error (misplaced fraction bar), and the central result (Eq. 46) is correct as independently verified. However, because Eq. (45) is load-bearing—it is the key intermediate step connecting the general moment expressions (Eq. 22) to the universal SR result (Eq. 46) and the exponential limit (Eq. 48)—the derivation chain is broken as written. A reader cannot verify the logic without independently re-deriving the K1→0 limit. The revision should be straightforward: correct Eq. (45), explicitly verify the chain to Eqs. (46) and (48), and add a sentence or two showing the key algebraic steps in the K1→0 limit. I do not see any deeper issues with the framework or the central claims."},"author_rebuttal":{"model":"glm-5.2","summary":"We thank the referee for a careful and constructive report. The referee correctly identifies a typographical error in Eq. (45) — a misplaced fraction bar — and asks us to verify the consistency of the corrected expressions with Eqs. (46)–(48). We agree with both points and will revise the manuscript accordingly. Below we address each comment in detail.","responses":[{"response":"The referee is entirely correct. Equation (45) contains a typographical error: a misplaced fraction bar causes the terms kBT/K and 2Dτ_off to appear under a common denominator, which is dimensionally inconsistent. The correct expressions are: ⟨x²⟩_{b,off} = kBT/K + 2Dτ_off(1/η_on − 1), ⟨x²⟩_{b,on} = kBT/K + 2Dτ_off/η_on. We have independently re-derived these from Eqs. (22) by carefully taking K₁ → 0 and confirm they match the referee's expressions. With the corrected forms, the difference ⟨x²⟩_{b,off} − ⟨x²⟩_{b,on} = −2Dτ_off is immediate, and substitution into Eq. (37) with K₁ → 0, K₂ = K yields ⟨Ẇ⟩ = (−K/2)(−2Dτ_off)/(τ_off + τ_on) = DKτ_off/(τ_off + τ_on), confirming Eq. (46). We also note that the incorrect version of Eq. (45) appeared only in the manuscript text; the numerical simulations shown in Fig. 2 were performed using the correct expressions, which is why they agree with Eq. (46). We will correct Eq. (45) in the revised manuscript.","revision_made":"yes","referee_comment":"Section IV, Eq. (45): The birth moments in the SR limit are dimensionally inconsistent. K has units [energy/length²] while Dτ_off has units [length²], so they cannot be added in a single denominator. The correct expressions are ⟨x²⟩_{b,off} = kBT/K + 2Dτ_off(1/η_on − 1) and ⟨x²⟩_{b,on} = kBT/K + 2Dτ_off/η_on. The error appears to be a misplaced fraction bar. As written, Eq. (45) does not reduce to the correct exponential limit (Eq. 48) and does not yield Eq. (46) when substituted into Eq. (37). The universal result ⟨Ẇ⟩ = DKτ_off/(τ_off + τ_on) is nonetheless correct."},{"response":"We agree and will add an explicit verification in the revised manuscript. The chain of reasoning is as follows. For exponential waiting times, η_on = τ_on/(τ_on + τ_K/2), and the second moment of the released-state residence time is ⟨a²⟩_{off} = 2τ_off². Starting from the corrected Eq. (45) and Eq. (47): ⟨x²⟩_{off} = ⟨x²⟩_{b,off} + (D/τ_off)⟨a²⟩_{off} = [kBT/K + 2Dτ_off(1/η_on − 1)] + 2Dτ_off = kBT/K + 2Dτ_off/η_on = ⟨x²⟩_{b,on}. This confirms the memoryless identity ⟨x²⟩_{off} = ⟨x²⟩_{b,on} stated in the text. Substituting η_on and using kBT/K = Dτ_K: ⟨x²⟩_{off} = Dτ_K + 2Dτ_off(τ_on + τ_K/2)/τ_on = kBT/K · (τ_off + τ_on)/τ_on + 2Dτ_off, which is exactly Eq. (48) for ⟨x²⟩_{off}. Similarly, ⟨x²⟩_{on} = kBT/K · (τ_off + τ_on)/τ_on from Eq. (47) directly, matching Eq. (48) for ⟨x²⟩_{on}. We will include this derivation, or a condensed version of it, in the revised Section IV to close the chain from the corrected Eq. (45) through Eq. (47) to Eq. (48).","revision_made":"yes","referee_comment":"Section IV, Eq. (47): Since Eq. (45) is incorrect as written, the consistency of Eq. (47) with Eq. (48) cannot be verified. The authors should explicitly show that the corrected Eq. (45), combined with Eq. (47), recovers Eq. (48) for exponential waiting times."}],"tokens_in":16468,"tokens_out":2915,"duration_ms":170524,"standing_objections":[]},"desk_editor":{"model":"glm-5.2","letter":"The main thing to know: this paper gets exact analytical results for a Brownian particle in a harmonic trap with arbitrary semi-Markovian switching, and the central result — universality of injected power in the stochastic-resetting limit — is correct. There is a typo in Eq. (45) that is worth flagging, but it does not invalidate the main conclusions. The paper deserves a serious referee who can check the algebra carefully and ask for the fix.","headline":"Paper derives exact results for semi-Markovian switching in a fluctuating harmonic trap; Eq. (45) has a typo that breaks the derivation chain but the central universality result survives.","tokens_in":17417,"tokens_out":979,"would_cite":true,"duration_ms":57154,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"glm-5.2","headline":"Injected power goes universal in a flickering trap","keywords":[],"falsifier":"If an experiment with a flickering optical trap and non-exponential switching schedules measured injected power that deviates from DKτ_off/(τ_off + τ_on) while mean on/off times are held fixed, the universality claim would be falsified.","tokens_in":16605,"feed_emoji":"🔄","tokens_out":1274,"duration_ms":104995,"temperature":0.7,"pith_summary":"The paper studies a Brownian particle in a harmonic trap whose stiffness switches between two values on a schedule drawn from arbitrary waiting-time distributions — not just exponential ones — making the dynamics semi-Markovian. The author restores a clean Markovian description by enlarging the state space to include the age of the current trap state (time since the last switch). In this framework, memory no longer lives in a nonlocal kernel but in the birth flux — the distribution of particle positions at the instant of each switch. Using this formulation, the author derives exact steady-state integral equations and closed-form second moments. The central discovery is a sharp split between energetic and spatial observables. The average potential energy is always kBT/2, independent of switching statistics, by a generalized virial relation. The injected power (rate of work done by the switching trap) generally depends on the full waiting-time distributions through a parameter A, with deterministic switching maximizing dissipation and intermittent switching suppressing it. But in the stochastic-resetting limit — where the trap alternates between off (free diffusion) and on (confinement) — the injected power simplifies to DKτ_off/(τ_off + τ_on), depending only on the mean on- and off-times and becoming completely insensitive to the shape of the waiting-time distributions. Meanwhile, spatial fluctuations in the released state retain explicit dependence on the second moment and asymptotic tails of the off-time distribution; heavy-tailed off-times can make the spatial variance diverge while the injected power stays finite and universal.","feed_headline":"Injected power goes universal in a flickering trap","feed_subtitle":"When a trap stochastically switches on and off, the energy cost depends only on average timings — but where the particle sits still feelsthe","key_machinery":"Age-structured Fokker-Planck equation; birth flux; self-consistent integral equations for birth and spatial distributions; variance-space deterministic switching dynamics","core_discovery":"In the stochastic-resetting limit of a semi-Markovian fluctuating harmonic trap, the injected power separates cleanly from spatial fluctuations: it depends only on mean residence times (universal), while the spatial distribution of the particle retains full sensitivity to higher moments and tails of the waiting-time distributions. This separation is made visible by the age-structured formulation, which encodes all memory in the birth flux — the distribution of positions at switching instants — rather than in a temporally nonlocal kernel. The author also introduces a variance-space reformulation in which the noisy spatial dynamics is replaced by deterministic transport of the variance, with a","pith_inferences":["If the universality of injected power extends beyond the two-state harmonic case, it could provide a general thermodynamic bound for intermittently confined systems where only mean residence times are known.","The variance-space formulation might generalize to higher-dimensional traps or anharmonic potentials where the variance dynamics is no longer closed, potentially revealing which observables retain universality and which do not.","The divergence hierarchy shift (confined state suppresses moment divergence by one order relative to released state) could serve as a diagnostic for identifying whether experimentally observed non-Gaussian tails originate from switching statistics or from other sources of anomalous diffusion."],"forward_implications":["Any experimental realization of stochastic resetting via trap switching can measure injected power without knowing the detailed switching statistics, since only mean on/off times are needed — simplifying thermodynamic accounting in single-particle experiments.","Heavy-tailed residence-time distributions in the released state produce algebraic spatial tails, offering a tunable mechanism for generating non-Gaussian stationary distributions in optical-trap experiments.","The variance-space reformulation reduces a noisy stochastic problem to deterministic transport with random switching, which could simplify numerical simulation and moment computation for broader classes of switching systems.","The finding that deterministic switching maximizes dissipation suggests a design principle: for fixed average switching frequency, regular protocols inject more energy than irregular ones."],"fun_headline_variants":["Injected power in a flickering trap separates cleanly from spatial memory","Energy cost in randomly switching traps depends only on mean timing","Age-structured formulation restores Markovianity for semi-Markovian traps","Switching harmonic trap dynamics decoupled via variance-space reformulation","Universal injected power emerges in stochastically resetting harmonic traps"],"cache_read_input_tokens":0,"weakest_assumption_plain":"The steady-state integral equations assume that the Ornstein-Uhlenbeck propagator fully captures spatial dynamics between switches and that a normalizable stationary density exists for arbitrary waiting-time distributions, but the conditions guaranteeing such normalizable solutions — especially for heavy-tailed distributions where spatial moments diverge — are not explicitly proven.","fun_headline_variants_meta":{"raw":{"variants":["Injected power in a flickering trap separates cleanly from spatial memory","Energy cost in randomly switching traps depends only on mean timing","Age-structured formulation restores Markovianity for semi-Markovian traps","Switching harmonic trap dynamics decoupled via variance-space reformulation","Universal injected power emerges in stochastically resetting harmonic traps"]},"model":"glm-5.2","effort":"high","cost_usd":0.0,"raw_usage":{"total_tokens":1105,"prompt_tokens":471,"completion_tokens":634,"prompt_tokens_details":null},"tokens_in":471,"tokens_out":634,"duration_ms":44626,"temperature":1.0,"reasoning_tokens":648,"cache_read_input_tokens":0,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-08T01:11:45.382817+00:00","model_set":{"reader":"glm-5.2"},"falsifier":"If an experiment with a flickering optical trap and non-exponential switching schedules measured injected power that deviates from DKτ_off/(τ_off + τ_on) while mean on/off times are held fixed, the universality claim would be falsified.","supporting_citations":[],"review_version":1}