{"id":"b20acf5d-3c7c-40b0-a989-83bbbfeecebd","arxiv_id":"2608.09205","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Photons selected by a small mask near an optical vortex have mean transverse momentum exceeding a bound set by local momentum conservation, providing experimental evidence for counterfactual violation in postselected quantum events.","lead":"This experiment shows that photons extracted from a tiny region near the core of an optical vortex can carry much more transverse momentum than any momentum present in the original beam, under a specific assumption about how the extraction mask behaves. The result is offered as the first laboratory evidence that individual quantum events can violate local conservation laws, a claim previously based only on thought experiments.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central claim stands or falls on assumption A1, yet the binary phase grating's recoil and the blocked-order momentum are never measured, leaving ordinary mask-induced momentum as an untested alternative.","rationale":"The paper is a careful experimental study of the superkick effect with large photon statistics and a parameter-free theoretical prediction that matches the measured transmitted distributions. The formal bound in Eq. (5) is a valid statistical maximum for any momentum-selection mechanism with fixed transmission probability, and the MLMC reconstruction is conservative in several respects (using theoretical Ptr, retaining pedestal in the full null, broadening K_M). My review of the logic confirms that the reader's weakest-assumption analysis is correct: the entire counterfactual-violation claim is conditional on A1. The theoretical support for A1 in Appendix B is a specific toy model (delta-in-time coupling, pointer degree of freedom) that has no demonstrated correspondence to a pixelated binary phase grating. The experiment does not measure the blocked diffraction orders or the mask recoil; the Discussion explicitly states that the compensating momentum change is not monitored. A local-momentum-conserving theory in which the binary grating imparts a small systematic momentum to the first-order beam—through phase errors, pixelation, or contrast-gradient effects—could reproduce the observed mean shift without any violation. This is not an accusation of error; it is an unverified physical premise. Because the paper is transparent about this conditionality and because the proposed checks are straightforward extensions of the existing apparatus, the verdict should remain CONDITIONAL rather than move to rejection.","tokens_in":27965,"tokens_out":23700,"duration_ms":330933,"concrete_test":"A direct check of A1 using existing data: compute the first moment of the measured mask-kick distribution K_M(q_y) from image MD4 (Gaussian through the mask). A nonzero centroid would falsify A1 and require shifting the MLMC null, likely erasing the claimed violation. A decisive new measurement: repeat the vortex run with the slit removed and image all diffraction orders (0, \\pm1, ...) simultaneously; if the probability-weighted total output-field momentum does not equal the input momentum, the mask has recoiled and A1 is false, whereas a balance supports A1.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's mean-momentum test compares the observed \\bar{q}_{T y} (Eq. 4, transmitted via the mask) with the tail bound \\bar{q}_{S y} (Eq. 5, maximum under momentum selection alone). The inequality |\\bar{q}_{T y}| \\le |\\bar{q}_{S y}| is a valid consequence of local momentum conservation only if assumption A1 holds: the mask does not supply the deterministic superkick. A1 is supported solely by the idealized photon-mask model in Appendix B, where the mask is a Gaussian amplitude mask with a delta-in-time interaction. The actual extraction device is a contrast-modulated binary phase grating on an SLM (Methods); it diffracts into multiple orders, only the first of which is detected. The experiment never measures the momentum of the complementary (blocked) orders or the mask's mechanical recoil, and the paper's Discussion explicitly concedes that the compensating momentum change is not monitored. If the grating's contrast modulation or pixelated phase structure imparts a systematic mean transverse momentum to the first-order beam beyond the constant grating offset that is removed in calibration, the observed \\bar{q}_{T y}=7.2\\pm0.3 (vs. bound 4.2\\pm0.1) could be ordinary momentum transfer, and no local-conservation violation would follow. Assumption A3 (pedestal exclusion) is a secondary, more model-dependent concern; A1 alone is sufficient to invalidate the headline claim if false.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports an experiment on photons in an optical vortex with charge ℓ=1, in which a small transmission mask extracts photons from the superoscillatory region near the vortex core. The authors derive an analytic expression for the mean transverse momentum of the transmitted photons, construct a 'maximal local momentum conserving' (MLMC) null distribution from the measured input momentum distribution and the measured mask-kick distribution, and find that the measured mean transverse momentum exceeds the MLMC bound by about five standard deviations, while the number of high-momentum photons exceeds the MLMC prediction by tens of standard deviations. The interpretation is that local momentum conservation is violated in the postselected subset of transmitted photons, conditionally on assumptions A1, A2, and A3 about the mask and about a weak nonintegrable pedestal in the input distribution. The theoretical derivation from Eq. (1) to Eq. (5) is transparent, the main comparison is parameter-free in the sense of having no fitted constant, and the photon-counting measurements are extensive. However, the central claim depends on untested assumptions about the physical mask, which is a binary phase grating rather than the idealized Gaussian amplitude mask used in the supporting model, and at least one central equation appears to be inconsistent with the numerical values reported in Table I.","tokens_in":28268,"tokens_out":19685,"duration_ms":189381,"significance":"If the interpretation is correct, this would be the first experimental evidence for the APR counterfactual violation of a local conservation law, with implications for the foundations of quantum mechanics and for possible nonlocal mechanisms restoring conservation. The paper is commendable for its clean parameter-free derivation of the bound, the explicit separation of assumptions A1-A3, the use of a conservative MLMC construction, and the high-statistics photon-counting data. The experimental excesses over the MLMC null are large and statistically unambiguous. The significance is nevertheless conditional: the headline claim stands or falls on whether the extraction mask imparts a systematic mean momentum that is not accounted for, and on whether the pedestal-exclusion assumption A3 is valid. These are physical premises about the measuring device, not statistical details, and the manuscript does not yet provide the measurements needed to support them.","major_comments":[{"comment":"The reported formula for the mean transmitted momentum is not the mean of the distribution f_T in Eq. (3). Direct integration of Eq. (3) for ℓ=1 gives qbar_Ty = ±(ℏ/d_M)[2 d_M^2(w0^2+w_M^2)]/[w_M^2(w0^2+w_M^2)+2w0^2 d_M^2]. For the parameters of Fig. 3(c), w_M=0.097w0 and d_M=0.10w0, this yields +6.84 ℏ/w0, matching the 'Theory' column of Table I, whereas Eq. (4) as printed gives about 14.9 ℏ/w0. The identical error appears in the Appendix A expression preceding Eq. (A.28) and contradicts the paper's own statement in Appendix A that qbar_Ty≈(2/3)q_s when w_M≈d_M (Eq. (4) gives q_s in that limit). Please correct Eq. (4), the corresponding Appendix A expression, and any statements that rely on them, and verify that the numerical values in Table I are derived from the corrected formula.","section":"Sec. II.A, Eq. (4); Appendix A, Eq. (A.27)"},{"comment":"The central inequality |qbar_Ty|≤|qbar_Sy| is derived only under assumption A1, that the deterministic superkick momentum is not supplied by the mask. The experimental mask is a contrast-modulated binary phase grating (Methods, Eq. C.2), not the Gaussian amplitude mask with delta-in-time interaction modeled in Appendix B. A phase grating diffracts into multiple orders, and the first detected order receives a fixed grating momentum that must be removed in calibration; the manuscript does not state how this offset is removed when fixing the momentum origin from the input LG distribution (Appendix D), nor does it report the mean momentum of the mask-transmitted Gaussian beam (MD4) relative to the unmasked Gaussian beam (MD3). The Appendix B model predicts zero mean mask recoil, but it does not describe the momentum balance among diffraction orders of a phase grating. Because the experiment never measures the blocked orders or the mask recoil, ordinary mask-induced mean momentum remains an untested alternative explanation for the observed excess over qbar_Sy. Please report the momentum balance in all diffraction orders, the measured mean of MD4, and an explicit justification of the offset subtraction before invoking A1.","section":"Sec. II.A (assumption A1), Appendix B, Methods (mask realization)"},{"comment":"Assumption A3, that the weak 1/|q_y| power-law pedestal is not preferentially selected by the mask, is load-bearing for the mean-momentum bound qbar_Sy and for the MLMC null. The pedestal subtraction is necessary because the pedestal is nonintegrable, but A3 is justified only by the physical statement that the apodization occurs far from the mask location. This is not directly tested: the mask could in principle select the corresponding spatial structure at the mask plane. Please provide a direct test of A3, for example by varying the aperture configuration and showing that qbar_Sy and the MLMC predictions are stable, or by characterizing the pedestal contribution in the plane of the mask and showing that it does not correlate with the mask transmission.","section":"Sec. II.B (assumption A3) and Methods/Appendix D (pedestal)"},{"comment":"The high-momentum excess test relies additionally on assumption A2, that the mask-kick distribution K_M is independent of the input photon momentum. The experimental MLMC is constructed by convolving the measured input tail f_S with K_M measured on a Gaussian input (MD4). If the grating's diffraction efficiency depends on incident angle or momentum, K_M for the high-momentum tail of the LG distribution could differ from K_M for the Gaussian input, and the reported excess of high-momentum photons would not be a valid rejection of the MLMC null. Please test A2 by measuring K_M for inputs displaced in momentum, or by directly comparing the transmitted distribution for a prepared high-q_y tail input with the convolution prediction.","section":"Sec. II.A (assumption A2) and Methods (experimental MLMC)"}],"minor_comments":[{"comment":"The abstract states that the bound derivation requires 'only the theoretically well-supported assumption' that the extraction mechanism does not alter the mean transverse momentum; given that the supporting Appendix B model is idealized and does not describe the actual phase grating, this wording overstates the current evidence for A1.","section":"Abstract"},{"comment":"The measured mask positions are reported as d_M=(0.10±0.01)w0 for panel (c) and d_M=(0.13±0.01)w0 for panel (f), while the nominal value is d_M=0.1w0; please clarify whether the difference is due to the measured vortex displacement and explain how this enters the theory values in Table I.","section":"Fig. 3 caption"},{"comment":"After correcting Eq. (4), please check that all 'Theory' entries in Table I are computed from the corrected formula and from the stated mask parameters in the figure caption; the current table is internally consistent with the corrected formula but not with the printed equation.","section":"Table I"},{"comment":"The phrase 'after correcting for the actual background based on a previous run of the same fit' is unclear; please specify exactly how the background correction was applied to the quoted pedestal parameters.","section":"Appendix D, Table II caption"},{"comment":"The uncertainty procedure assigns a constant uncertainty per point by requiring reduced chi-square equal to unity; a brief statement that this procedure captures known optical imperfections without double-counting the pedestal Monte Carlo uncertainty would improve reproducibility.","section":"Methods, Uncertainty estimation"}],"recommendation":"major_revision","confidential_remarks":"This is a well-executed and thought-provoking experiment, and the referee recognizes the novelty of the claim. The main issue is not the statistical significance but the physical interpretation: assumption A1 is unsupported for the actual phase grating, and the experiment does not measure the compensating channels. The incorrect Eq. (4) also needs fixing. I would recommend major revision with a request for the additional mask-characterization measurements described in the major comments; if the authors can provide those, the paper could be suitable for publication in a high-profile venue."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper to know about: Cameron et al. report an experiment where photons are extracted from the superoscillatory region near the core of an optical vortex, and the mean transverse momentum of the transmitted photons is 7.2 ± 0.3 in units of hbar/w0, against a bound of 4.2 ± 0.1 derived from a local-momentum-conserving null model. The derivation from Eq. (1) through Eq. (4) is clean and parameter-free, the MLMC null is constructed from measured input and mask-kick distributions, and the excess is tens of standard deviations. The paper also does something rare: it states its assumptions explicitly and labels them A1, A2, A3. The statistical work is careful, and the pedestal handling is conservative.\n\nThe soft spot is A1: the claim that the deterministic superkick is not supplied by the extraction mask. The supporting model in Appendix B treats the mask as a Gaussian amplitude screen with a delta-in-time interaction. The actual device is a contrast-modulated binary phase grating on an SLM. The first diffraction order is selected; the complementary orders are never measured. The experiment does not measure the mask's mechanical recoil, and the Discussion concedes that the compensating momentum change is not monitored. If the grating's pixelated phase structure or contrast modulation imparts a systematic mean transverse momentum to the first-order beam beyond the constant offset removed in calibration, the observed excess is ordinary momentum transfer and no conservation-law violation follows. That is not a far-fetched scenario; it is an untested physical premise about the measuring device.\n\nNote that the paper is honest about this: the bound is derived under assumption A1, and the abstract says the assumption is 'well-supported.' The support is a theoretical model that may not describe the actual grating. A3 (pedestal exclusion) is a secondary, more model-dependent concern; A1 alone is enough to sink the interpretation if false.\n\nWho is this for? Quantum foundations and structured-light optics readers. The experimental technique and the MLMC construction are worth engaging with regardless of the interpretation. If I were refereeing, I would recommend publication only after the mask-recoil question is addressed—either by a direct measurement of the complementary orders or a model that accounts for the actual grating structure. But the paper deserves that referee time. It is a serious, internally consistent piece of work, not a crank claim. My verdict: engage, but treat the headline as conditional.","headline":"Strong, careful optics experiment with a clean null model, but the headline claim of local conservation violation rests on an untested assumption about the mask's recoil; still deserves serious review.","tokens_in":28775,"tokens_out":2692,"would_cite":true,"duration_ms":25928,"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 narrow mask near an optical vortex core extracts photons whose mean transverse momentum exceeds the maximum allowed by local momentum conservation, the paper reports.","keywords":["superoscillations","optical vortex","local momentum conservation","superkick","postselection","counterfactual violation","Laguerre-Gauss modes","quantum foundations"],"falsifier":"Mount the actual binary phase grating on a momentum-sensitive mechanical support and measure its center-of-mass recoil while photons are transmitted; if the mask acquires a mean momentum equal and opposite to the observed superkick $\\bar{q}_{Ty}$, assumption A1 fails and local momentum conservation is restored.","tokens_in":27752,"feed_emoji":"🌀","tokens_out":6546,"duration_ms":67089,"temperature":0.7,"pith_summary":"The paper reports an experiment designed to test a claim of Aharonov, Popescu, and Rohrlich: that individual quantum events can violate local conservation laws even though ensemble averages obey them. Photons in a Laguerre-Gauss vortex are passed through a small Gaussian transmission mask placed near the vortex core, where the field is superoscillatory. The transmitted photons have a mean transverse momentum of about $7.2\\,\\hbar/w_0$ in one dataset, compared with a maximum of about $4.2\\,\\hbar/w_0$ allowed by any local-momentum-conserving theory under the stated assumptions; the paper reports a discrepancy above five standard deviations. If correct, this is the first experimental evidence that local momentum conservation can fail in postselected quantum events, with the violation arising only counterfactually and hence not conflicting with relativistic causality.","feed_headline":"Vortex-core photons exceed the local momentum limit","feed_subtitle":"A mask-selected subset of photons carries more sideways momentum than any local-conserving selection could allow.","key_machinery":"The operative object is the superoscillatory region near the core of a Laguerre-Gauss vortex, where the local phase gradient, and hence local transverse momentum, can exceed every momentum in the Fourier spectrum. Extraction is by a Gaussian transmission mask of width $w_M$ centered at distance $d_M$ from the core, for which the mean transmitted momentum is $\\bar{q}_{Ty}=\\pm\\frac{\\hbar}{d_M}\\left(\\frac{w_0^2}{w_0^2+w_M^2}+\\frac{w_M^2}{2d_M^2}\\right)$, the superkick. The reference null hypothesis is the maximal local-momentum-conserving (MLMC) distribution: take the upper or lower tail of the input momentum distribution whose probability equals the mask transmission, then convolve it with the mask-kick distribution, producing the most favorable local-conserving hypothesis. The argument rests on assumptions A1 (no systematic mask recoil), A2 (mask kick independent of input momentum), and A3 (the weak apodization pedestal is not preferentially selected by the mask).","core_discovery":"The discovery the paper seeks to establish is that local momentum conservation is violated in the conditional subset of quantum events in which a photon is extracted from a superoscillatory region of an optical vortex. The argument is statistical: the measured mean transverse momentum of mask-transmitted photons lies beyond the bound $\\bar{q}_{Ty}\\le\\bar{q}_{Sy}$ derived from the input momentum distribution under the sole assumption that the mask does not supply the superkick momentum, and the measured number of high-momentum photons exceeds the maximally biased local-conserving reference distribution. The authors emphasize that the violation is counterfactual: input and transmitted momentum distributions are measured in mutually exclusive configurations, and the inference uses the prepared vortex state as the counterfactual input. They also state that the total ensemble, including blocked photons, conserves momentum, so the violation is conditional and does not imply that global conservation fails.","pith_inferences":["If the reported violation holds, conservation laws at the level of individual events would be statistical rather than strict, which would change how measurement back-action is budgeted in quantum optics and weak-measurement setups.","A natural follow-up, suggested by the authors' discussion but not performed here, is to test the compensation mechanism with two entangled photons and look for a momentum anti-correlation that restores total-momentum conservation in each run.","The counterfactual inference used here is weaker than Bell-test local realism; generalizing it to other conserved quantities and other superoscillatory platforms would show whether this is a generic feature of postselected quantum events or specific to vortex masks."],"forward_implications":["Local momentum conservation would hold only as an ensemble statement, not for individual postselected events, so quantum back-action cannot always be balanced locally event by event.","The observed excess of high-momentum photons beyond the MLMC reference provides a second, distribution-level test of the same violation.","Because the comparison is counterfactual, the experiment does not permit superluminal signaling, and a delayed-choice version could make the measurement setting spacelike separated from the preparation.","Including blocked photons restores average momentum conservation, so any observable violation is confined to the rare transmitted subset of quantum events."],"supporting_citations":[{"why":"Supplies the APR thought-experiment argument that individual quantum events can violate local conservation laws even when ensemble averages conserve them.","marker":"[18]"},{"why":"Provides the entanglement-based compensation mechanism and the conservation-law analysis that the paper's Appendix B adapts to model mask recoil.","marker":"[19]"},{"why":"Defines the superkick and the multiple local-momentum definitions near an optical vortex, including the weak-value momentum used to interpret the result.","marker":"[22]"},{"why":"Predicts superweak momentum transfer near optical vortices, the atomic-physics precursor that the photon-mask experiment is designed to realize.","marker":"[23]"},{"why":"Identifies vortex cores as superoscillatory regions, grounding the claim that the mask extracts photons from a superoscillatory part of the wavefunction.","marker":"[21]"},{"why":"Refines the accounting of conservation laws for every quantum measurement outcome and supports the photon-mask interaction model in Appendix B.","marker":"[25]"},{"why":"Provides the holographic method used to prepare the input Laguerre-Gauss vortex mode on the spatial light modulator.","marker":"[27]"}],"fun_headline_variants":["Superoscillating photons break local momentum conservation","Vortex photons violate local momentum bound","Counterfactual quantum kick: photons exceed local limit","Local momentum conservation fails for vortex-core photons"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The violation claim depends on assumption A1: the deterministic superkick momentum is not supplied by the mask itself; if the actual binary grating recoils with that momentum, the observed mean shift would be ordinary momentum transfer and the central claim collapses.","fun_headline_variants_meta":{"raw":{"variants":["Superoscillating photons break local momentum conservation","Vortex photons violate local momentum bound","Counterfactual quantum kick: photons exceed local limit","Local momentum conservation fails for vortex-core photons"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00035,"raw_usage":{"total_tokens":1905,"prompt_tokens":931,"completion_tokens":974,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":547,"completion_tokens_details":{"reasoning_tokens":917}},"tokens_in":547,"tokens_out":974,"duration_ms":8262,"temperature":1.0,"reasoning_tokens":917,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T21:24:34.726573+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Mount the actual binary phase grating on a momentum-sensitive mechanical support and measure its center-of-mass recoil while photons are transmitted; if the mask acquires a mean momentum equal and opposite to the observed superkick $\\bar{q}_{Ty}$, assumption A1 fails and local momentum conservation is restored.","supporting_citations":[{"cited_title":"Aharonov, S","cited_arxiv_id":null,"evidence_quote":"Supplies the APR thought-experiment argument that individual quantum events can violate local conservation laws even when ensemble averages conserve them."},{"cited_title":"Aharonov, S","cited_arxiv_id":null,"evidence_quote":"Provides the entanglement-based compensation mechanism and the conservation-law analysis that the paper's Appendix B adapts to model mask recoil."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines the superkick and the multiple local-momentum definitions near an optical vortex, including the weak-value momentum used to interpret the result."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Predicts superweak momentum transfer near optical vortices, the atomic-physics precursor that the photon-mask experiment is designed to realize."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Identifies vortex cores as superoscillatory regions, grounding the claim that the mask extracts photons from a superoscillatory part of the wavefunction."},{"cited_title":"Collins and S","cited_arxiv_id":null,"evidence_quote":"Refines the accounting of conservation laws for every quantum measurement outcome and supports the photon-mask interaction model in Appendix B."},{"cited_title":"Bolduc, N","cited_arxiv_id":null,"evidence_quote":"Provides the holographic method used to prepare the input Laguerre-Gauss vortex mode on the spatial light modulator."}],"review_version":1}