REVIEW 2 major objections 8 minor 2 cited by
$H \to ZZ$ as a double-slit experiment
T0 review · 2 major / 8 minor · reviewed 2026-08-12 · deepseek-v4-flash
Pith's one-line read This paper argues that the Higgs decay $H \to ZZ \to 4\ell$ can be viewed as a double-slit experiment, with the $ee\mu\mu$ final state playing the role of 'one slit covered' and the $4e/4\mu$ final states the role of 'both slits open,' so…
desk verdict Clever double-slit mapping and clean parton-level numbers, but the HL-LHC sensitivity projection relies on a symmetry baseline that the selection cuts break; needs an interference-free control before the 4-5 sigma claim is credible. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The machinery is the four-dimensional decay angular distribution of leptonic $Z$ decays expanded in spherical harmonics, Eq. (1), and in particular the coefficients $c_{111-1}$ and $c_{1010}$, which are rank-2 spin correlations suppressed by the factor $\eta_\ell \simeq 0.13$ in $ee\mu\mu$. In the distinguishable $ee\mu\mu$ final state, these coefficients can be evaluated for the correct flavour pairing, the wrong pairing, and the 'both pairings' average; the latter is an incoherent sum. In $4e/4\mu$ final states, summing over the two possible pairings does not yield the same average, and the deviation from the $ee\mu\mu$ average is the signature of interference.
What would settle it
Measure $c_{111-1}$ and $c_{1010}$ in $ee\mu\mu$ and $4e/4\mu$ final states at the HL-LHC using the same pairing-blind selection, with the upper cut $m_{Z1} \le M_Z$ as in the paper. Under the no-interference hypothesis, the $4e/4\mu$ values should equal the average of the two $ee\mu\mu$ flavour pairings; under interference, one expects $c_{111-1} \approx -0.03$ and $c_{1010} \approx 6.23$ at parton level. A measurement consistent with the average, or a difference that changes sign when muon and electron acceptance cuts are interchanged, would falsify the claim.
Extended reading notes
Core claim
The central discovery is that in $4e/4\mu$ final states the angular correlation coefficients $c_{111-1}$ and $c_{1010}$, obtained by counting each event with both lepton pairings, are not the average of the two $ee\mu\mu$ flavour pairings. At parton level, $ee\mu\mu$ with both pairings gives $c_{111-1} = -0.556$ and $c_{1010} = 7.093$, whereas $4e/4\mu$ with both pairings gives $c_{111-1} = -0.032$ and $c_{1010} = 6.232$. The difference is interpreted as a coherent interference between the two Feynman diagrams related by identical-particle exchange, analogous to the double-slit experiment. Using pseudo-experiments with reconstructed events, the paper projects that the combined statistical significance of the difference between $ee\mu\mu$ and $4e/4\mu$ could reach $4.9\sigma$, and the comparison with $4\mu$ alone $4.5\sigma$, at the HL-LHC.
Load-bearing premise
The no-interference baseline is assumed to be the average of the two $ee\mu\mu$ pairings, justified by a symmetry argument that is not written out; if that baseline is wrong, or if electron/muon acceptance differences mimic the effect, the interference conclusion fails.
Editorial extensions
If this is right
- If correct, a purely data-driven comparison of $ee\mu\mu$ and $4e/4\mu$ final states can establish quantum interference without relying on Monte Carlo predictions for the signal shape.
- It would provide the first experimental evidence that muons behave as identical fermions, a property never directly tested before.
- At the HL-LHC, the combined significance from $c_{111-1}$ and $c_{1010}$ should reach about $4$ to $5$ standard deviations, enough to claim observation in particle physics.
- The electroweak background is small and has similar coefficients in the two channels, so background subtraction is feasible and would not erase the effect.
- Other terms in the angular distribution, such as $a_{20}$, could be combined to further improve sensitivity.
Reading between the lines
- One could transfer the pairing-blind analysis to $H \to WW$ or to four-charged-lepton final states at future lepton colliders, where larger statistics and cleaner flavor assignments may sharpen the test.
- The experiment could be recast as a which-way measurement: instead of comparing final states, one could tag one of the two $Z$ bosons by a recoiling system, turning the double-slit into a which-slit measurement and checking that the interference pattern disappears.
- If the predicted difference between $ee\mu\mu$ and $4e/4\mu$ is not seen at the HL-LHC, that would point either to an acceptance bias mimicking the $ee\mu\mu$ baseline or to a genuine breakdown of the identical-particle exchange assumption, in either case a notable result.
- Explicitly deriving the symmetry baseline (currently asserted rather than written out) would allow one to predict the interference term analytically rather than by Monte Carlo.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. This paper proposes a test of quantum interference and identical-particle effects in H→ZZ→4ℓ decays by comparing angular correlation coefficients c111−1 and c1010 extracted from eeµµ and 4e/4µ final states. In eeµµ, two pairings (same-flavour 'A' and different-flavour 'B') can be used; counting each event twice yields the average of pairings A and B. In 4e/4µ, both pairings are included coherently, and the paper argues that the resulting coefficients differ from the eeµµ average, providing an analogue of the double-slit experiment. At parton level, c111−1 = −0.556 for eeµµ both pairings versus −0.032 for 4e/4µ, while c1010 = 7.093 versus 6.232. Detector-level simulations with Delphes for an HL-LHC luminosity of 3 ab−1 give projected significances of 2.9σ and 3.5σ for the two coefficients, combined to 4.9σ, and 4.5σ for the eeµµ versus 4µ comparison. The paper also proposes mass and BDT pairing criteria to test identical-particle effects.
Significance. If established, this would be a novel, data-driven probe of quantum interference at high energies and the first test of identical-particle behaviour of muons. The work has strong practical elements: the Monte Carlo demonstration is internally consistent (the both-pairings results are exactly the averages of the individual pairings), and the pseudo-experiment procedure for statistical uncertainties is well defined. The proposed comparison is falsifiable and could be carried out with existing tools. However, the central interpretive claim depends on an as-yet unvalidated no-interference baseline, and the sensitivity projection does not include background subtraction or systematic uncertainties; these issues must be resolved before the claimed significance can be taken at face value.
major comments (2)
- [Probing quantum interference, after Eq. (5)] The assertion that in the absence of interference the 4e/4µ both-pairings coefficients equal the average of the eeµµ pairings A and B is not derived; the text only says 'as it can be seen by symmetry arguments.' This identity is load-bearing because the entire interference claim is the difference between the 4e/4µ value and this average. The two pairings in 4e/4µ are both same-flavour, while pairing B in eeµµ is a different-flavour pairing, so the mapping between the two is not self-evident. Furthermore, the detector-level analysis in 'Future prospects' applies flavour-dependent acceptance cuts (electrons |η|≤2.5, pT≥7 GeV; muons |η|≤2.7, pT≥5 GeV) and an upper cut mZ1≤MZ, which explicitly break the electron–muon exchange symmetry required for the identity. Reconstructed values in Table I show eeµµ c111−1 shifting from −0.556 at parton level to −0.76 with both pairings, while 4e/4µ remains near −0.03, i.e., acceptance and cuts increase the apparent separation in the direction expected if the baseline is broken. The authors should provide an explicit no-interference baseline, for example by computing the 4e/4µ cross section with the interference term removed in the matrix element, or by applying a flavour-blind pairing to eeµµ events after the same detector selection; until this is done, the difference between channels cannot be unambiguously attributed to quantum interference.
- [Future prospects, Table II and background paragraph] The pseudo-experiments for the statistical significances are drawn from the expected number of events listed in the text (1144, 379, 1106), which appears to be the signal yield only, without including the electroweak background whose size is about 1/4 of the signal. Table I shows that the background coefficients differ between eeµµ and 4e/4µ (e.g., c111−1 bkg = 1.45 vs 1.42; c1010 bkg = 7.67 vs 7.68), so an analysis that does not subtract the background will measure a mixture. The paper states that background subtraction is out of scope and will cause a 'mild' decrease of significance, but this is not quantified. Given that the central claim is a projected 4.9σ effect, the projection should be repeated with background included in the pseudo-experiments, with a validated subtraction procedure, and with an estimate of dominant systematic uncertainties (e.g., lepton efficiencies, energy calibration, and the shape of the mZ1 distribution).
minor comments (8)
- [Abstract and text] The notation '4e/4µ' is ambiguous because it could be read as a single 2e2µ final state; since the paper means '4e or 4µ', please define this shorthand explicitly at first use.
- [Eqs. (4) and (5)] The parton-level coefficients are quoted without Monte Carlo statistical uncertainties; adding them would help the reader judge the significance of the parton-level discrepancy before detector simulation.
- [Future prospects] The sentence 'We require two opposite-sign same-flavour leptons' is incomplete; please clarify that the four-lepton selection requires two same-flavour opposite-sign pairs.
- [Future prospects] The text gives the expected number of events as 1144, 379 and 1106 but does not state whether these are the signal-only yields in the mass window 120≤m4ℓ≤130 GeV and whether they include the 4e and 4µ branching fractions; please specify the exact definition.
- [Probing identical-particle effects] The statement that the low-mZ1 region where mass pairing fails 'amounts to a small fraction of the decay width' should be quantified, since the fraction determines the accuracy of the mass-pairing proxy for the true pairing.
- [Footnote 1] The claim that coefficients for L=3 are compatible with zero within Monte Carlo uncertainty is not quantified; please state the sample size and the uncertainty on the relevant coefficients.
- [Reference [9]] Reference [9] contains a typo ('System´´' for 'System') and the arXiv category is hep-ph while the reference is a computer science proceedings paper; please check the bibliographic data.
- [Table II] The header 'Z' in the last column is undefined; please state that it is the statistical significance in units of standard deviations and describe how it is computed from the pseudo-experiment distributions.
Circularity Check
No significant circularity: the eeµµ-vs-4e/4µ comparison is computed from first-principles MC with no fit to the claimed interference signal.
full rationale
The central claim (4e/4µ both-pairings coefficients differ from the eeµµ both-pairings average) is an output of a Monte Carlo calculation, not a fit. The paper uses the angular-coefficient formalism from the author's earlier work [7,8], but the numerical values for c111−1 and c1010 are obtained by generating H→ZZ→4ℓ at LO with MadGraph in full phase space (Eqs. (4)-(5)) and with Delphes detector simulation for the projections (Tables I-II). The 'absence of interference' baseline is justified by a symmetry argument; even if that argument is underived and flavour-dependent acceptance could affect it, this is a correctness/robustness concern and not a circular reduction: the baseline is not defined in terms of the 4e/4µ result, and no parameter is adjusted to force the quoted 2.9σ/3.5σ/4.9σ differences. The self-citations [7,8] provide the spin-density-matrix formalism but carry no weight in establishing the numerical interference effect, which is independently computed. Therefore no circularity is present.
Assumptions & free parameters
free parameters (1)
- Upper kinematic cut mZ1 ≤ MZ =
MZ ≈ 91.2 GeV
assumptions (5)
- domain assumption The H→ZZ→4ℓ amplitude is dominated by two tree-level diagrams related by exchange of identical leptons.
- domain assumption The angular distribution in Eq. (1), truncated at L=2, fully describes both eeμμ and 4e/4μ final states.
- ad hoc to paper In the absence of interference, the 4e/4μ both-pairings result would equal the average of the two eeμμ pairings.
- domain assumption Leading-order Monte Carlo with Pythia and Delphes simulation gives an adequate description of the signal and background for the sensitivity projection.
- domain assumption The electroweak four-lepton background is small and has similar correlation coefficients in both channels, so subtracting it would not change the qualitative conclusion.
Cite this review
Pith. "Pith review of $H \to ZZ$ as a double-slit experiment." pith.science (2026). https://pith.science/paper/LAJYDUZR
@misc{pith2026241113464,
author = {Pith},
title = {Pith review of: $H \to ZZ$ as a double-slit experiment},
year = {2026},
howpublished = {\url{https://pith.science/paper/LAJYDUZR}},
note = {Machine review of arXiv:2411.13464}
}
abstract
The decay $H \to ZZ \to 4\ell$, with $\ell = e,\mu$, can be used to test quantum interference in analogy to the famous double-slit experiment. The observations corresponding to `covering a slit' can be extracted from data in the $ee\mu\mu$ channel, while the observations when both `slits' are open, which include the interference, are measured in the $4e/4\mu$ channels. In this way, this process offers a unique opportunity to investigate identical-particle effects at high energies, and provide the first evidence of identical-particle behaviour for muons. Sensitivity at the $4\sigma$ level could be achieved at the high-luminosity upgrade of the Large Hadron Collider.
Figures
Forward citations
Cited by 2 Pith papers
-
Qubit-qubit-qutrit quantum correlations in $H \to f \bar f V$
In h→τ^-τ^+ Z decays, the spin state is genuinely qubit-qubit-qutrit entangled almost everywhere, violates Bell inequalities throughout, and carries up to 1.95 bits of non-local magic.
-
Quantum Tomography and Entanglement in Semi-Leptonic $h\to VV^*$ Decays at Higher Orders
Semi-leptonic h→VV* decays retain an effective two-qutrit quantum description under NLO QCD and electroweak corrections, unlike the fully leptonic h→4ℓ channel.
Reference graph
Works this paper leans on
-
[1]
The other boson is labelled asZ2, with decay productsℓ+ 2,ℓ − 2. •ˆris defined asˆr=sign(cosθ)(ˆp p−cosθ ˆk)/sinθ, withˆpp =(0,0,1)the direction of one proton in the laboratory frame, andcosθ= ˆk⋅ˆpp. •ˆnis taken orthogonal,ˆn= ˆk׈r. The angular orientation of the decay products can be specified by the polar and azimuthal anglesΩ1=(θ 1, ϕ1) andΩ 2 =(θ 2,...
arXiv 2025
-
[2]
And, as before, label asZ1 the one with largest invariant mass. The reference system is then con- structed as described for distinguishable particles, and likewise the four-dimensional decay distribution has the form (1).1 However, the numerical values of the coef- ficientsa 1,2 LM andc L1M1L2M2 differ from theeeµµfinal state. In particular,c 111−1 andc 1...
-
[3]
Observation of quan- tum entanglement with top quarks at the AT- LAS detector,
G. Aadet al.[ATLAS], “Observation of quan- tum entanglement with top quarks at the AT- LAS detector,” Nature633(2024) no.8030, 542-547 [arXiv:2311.07288 [hep-ex]]. Briefing in https://atlas.cern/Updates/Briefing/Top- Entanglement
arXiv 2024
-
[4]
A. Hayrapetyanet al.[CMS], “Observation of quantum entanglement in top quark pair pro- duction in proton–proton collisions at √s= 13TeV,” Rept. Prog. Phys.87(2024) no.11, 117801 [arXiv:2406.03976 [hep-ex]]. Briefing inhttps://cms.cern/news/entangled-titans- unravelingmysteries-quantummechanics-top-quarks
arXiv 2024
-
[5]
A. Hayrapetyanet al.[CMS], “Measurements of polarization and spin correlation and observation of entanglement in top quark pairs using lep- ton+jets events from proton-proton collisions at √s = 13 TeV,” [arXiv:2409.11067 [hep-ex]]. Briefing inhttps://cms.cern/news/spooky-action-distance- between-heaviest-particles
-
[6]
Measurement of subpicosecond time intervals between two photons by interference,
C. K. Hong, Z. Y. Ou and L. Mandel, “Measurement of subpicosecond time intervals between two photons by interference,” Phys. Rev. Lett.59(1987), 2044-2046
work page 1987
-
[7]
Electron- Electron and Positron-Electron Scattering Measure- ments
A. Ashkin, L. A. Page and W. M. Woodward, “Electron- Electron and Positron-Electron Scattering Measure- ments” Phys. Rev.94(1954), 357–362
work page 1954
-
[8]
W. Bernreuther, D. Heisler and Z. G. Si, “A set of top quark spin correlation and polarization observables for the LHC: Standard Model predictions and new physics contributions,” JHEP12(2015), 026 [arXiv:1508.05271 [hep-ph]]
arXiv 2015
Show all 22 references
-
[9]
Testing entanglement and Bell inequali- ties inH→ZZ,
J. A. Aguilar-Saavedra, A. Bernal, J. A. Casas and J. M. Moreno, “Testing entanglement and Bell inequali- ties inH→ZZ,” Phys. Rev. D107(2023) no.1, 016012 [arXiv:2209.13441 [hep-ph]]
2023 arXiv
-
[10]
TheZboson spin observables as messengers of new physics,
J. A. Aguilar-Saavedra, J. Bernabéu, V. A. Mitsou and A. Segarra, “TheZboson spin observables as messengers of new physics,” Eur. Phys. J. C77(2017) no.4, 234 [arXiv:1701.03115 [hep-ph]]. 6
2017 arXiv
-
[11]
Chen and C
T. Chen and C. Guestrin, “XGBoost: A Scalable Tree Boosting System´´Proceedings of the 22nd ACM SIGKDD International Conference on Knowl- edge Discovery and Data Mining (KDD’16), 785 (2016) [arXiv:1603.02754 [hep-ph]]
2016 arXiv
-
[12]
The automated computation of tree-level and next-to-leading order differential cross sections, and their matching to parton shower simulations,
J. Alwall, R. Frederix, S. Frixione, V. Hirschi, F. Mal- toni, O. Mattelaer, H. S. Shao, T. Stelzer, P. Torrielli and M. Zaro, “The automated computation of tree-level and next-to-leading order differential cross sections, and their matching to parton shower simulations,” JHEP...
2014 arXiv
-
[13]
Report from Working Group 2: Higgs Physics at the HL-LHC and HE-LHC,
M. Cepeda, S. Gori, P. Ilten, M. Kado, F. Riva, R. Ab- dul Khalek, A. Aboubrahim, J. Alimena, S. Alioli and A. Alves,et al.“Report from Working Group 2: Higgs Physics at the HL-LHC and HE-LHC,” CERN Yellow Rep. Monogr.7(2019), 221-584 [arXiv:1902.00134 [hep- ph]]
2019 arXiv
-
[14]
Handbook of LHC Higgs Cross Sec- tions: 4. Deciphering the Nature of the Higgs Sector,
D. de Florianet al.[LHC Higgs Cross Section Work- ing Group], “Handbook of LHC Higgs Cross Sec- tions: 4. Deciphering the Nature of the Higgs Sector,” arXiv:1610.07922 [hep-ph]
-
[15]
A Brief In- troduction to PYTHIA 8.1,
T. Sjostrand, S. Mrenna and P. Z. Skands, “A Brief In- troduction to PYTHIA 8.1,” Comput. Phys. Commun. 178(2008), 852-867 [arXiv:0710.3820 [hep-ph]]
2008 arXiv
-
[16]
DELPHES 3, A modular framework for fast simulation of a generic col- lider experiment,
J. de Favereauet al.[DELPHES 3], “DELPHES 3, A modular framework for fast simulation of a generic col- lider experiment,” JHEP02(2014), 057 [arXiv:1307.6346 [hep-ex]]
2014 arXiv
-
[17]
Test of CP-invariance of the Higgs boson in vector-boson fusion production and in its decay into four leptons,
G. Aadet al.[ATLAS], “Test of CP-invariance of the Higgs boson in vector-boson fusion production and in its decay into four leptons,” JHEP05(2024), 105 [arXiv:2304.09612 [hep-ex]]
2024 arXiv
-
[18]
Measurements of produc- tion cross sections of the Higgs boson in the four-lepton final state in proton–proton collisions at√s=13TeV,
A. M. Sirunyanet al.[CMS], “Measurements of produc- tion cross sections of the Higgs boson in the four-lepton final state in proton–proton collisions at√s=13TeV,” Eur. Phys. J. C81(2021) no.6, 488 [arXiv:2103.04956 [hep-ex]]
2021 arXiv
-
[19]
Sulla determinazione empirica di una legge di distribuzione,
A. N. Kolmogorov, “Sulla determinazione empirica di una legge di distribuzione,” Giornale dell’Istituto Italiano degli Attuari, 4, 83-91 (1933)
1933
-
[20]
N. V. Smirnov, ”Table for Estimating the Goodness of Fit of Empirical Distributions,” The Annals of Mathematical Statistics, 19(2), 279-281 (1948)
1948
-
[21]
From angular coefficients to quantum observables: a phenomenological appraisal in di-boson systems,
M. Grossi, G. Pelliccioli and A. Vicini, “From angular coefficients to quantum observables: a phenomenological appraisal in di-boson systems,” [arXiv:2409.16731 [hep- ph]]
-
[22]
Measurement of the impact-parameter dependent azimuthal anisotropy in coherentρ 0 photoproduction in Pb–Pb colli- sions at √sN N =5.02TeV,
S. Acharyaet al.[ALICE], “Measurement of the impact-parameter dependent azimuthal anisotropy in coherentρ 0 photoproduction in Pb–Pb colli- sions at √sN N =5.02TeV,” Phys. Lett. B858 (2024), 139017 [arXiv:2405.14525 [nucl-ex]]. Briefing in https://home.cern/news/news/physics/a...
2024 arXiv
Reviewed August 12, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.