REVIEW 3 major objections 5 minor 1 cited by
Search for an Anomalous Production of Charged-Current $\nu_e$ Interactions Without Visible Pions Across Multiple Kinematic Observables in MicroBooNE
T0 review · 3 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash
Pith's one-line read MicroBooNE's complete five-year dataset excludes an electron-like interpretation of the MiniBooNE low-energy excess at more than 99% confidence in every kinematic variable tested.
desk verdict A careful, incremental but important update that excludes two specific electron-like models of the MiniBooNE excess at >99% CL_s, with the abstract slightly overstating the breadth of that exclusion. 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 analysis is built on two signal selections — $1e\mathrm{N}p0\pi$ (at least one proton with kinetic energy above 40 MeV) and $1e0p0\pi$ (no visible protons) — both requiring a single electron-like shower and no pions. The discriminating machinery is a sideband-constraint procedure that uses high-statistics $\nu_\mu$ control channels and a neutral-pion control sample, combined through the block matrix method, to reduce systematic uncertainty on the signal prediction by roughly 40%. The two low-energy-excess (LEE) signal models are the central objects: Model 1 unfolds the MiniBooNE excess as a function of reconstructed neutrino energy using a smearing matrix for charged-current quasi-elastic events, while the new Model 2 unfolds the excess in the two-dimensional space of reconstructed shower energy and $\cos\theta$ and applies the resulting scale factor to MicroBooNE's intrinsic $\nu_e$ prediction without altering hadronic kinematics. Statistical conclusions come from frequentist pseudo-data trials, confidence intervals on a fitted signal strength, and the modified frequentist $\mathrm{CL}_s$ method for two-hypothesis rejection.
What would settle it
A concrete test: if a reprocessed MiniBooNE unfolding with updated efficiency and smearing produced an excess whose electron-energy and angle shapes, injected into MicroBooNE's simulation with full signal-model systematics, significantly improved the $\chi^2$ of the $1e0p0\pi$ channel relative to the null hypothesis, the paper's central exclusion would be contradicted. Alternatively, showing that the $1e\mathrm{N}p0\pi$ deficit vanishes under a corrected argon cross-section model while a Model-2-like excess appears in the data would overturn the conclusion.
Extended reading notes
Core claim
On the paper's own terms, the central discovery is that no anomalous excess of charged-current $\nu_e$ interactions without visible pions appears in MicroBooNE's full dataset. The observed event counts are consistent with the nominal standard-model prediction under the null hypothesis, with frequentist $p$-values of 32%, 27%, and 44% for reconstructed neutrino energy, shower energy, and shower $\cos\theta$, respectively, for the combined channels. When the MiniBooNE excess is injected as a signal through either of two empirical models, the data prefer the null hypothesis, and the modified frequentist $\mathrm{CL}_s$ method excludes the electron-like interpretation at $>99\%$ confidence in all variables. The exclusion is driven mainly by the $1e\mathrm{N}p0\pi$ channel, which shows a 24% deficit relative to prediction (2.4$\sigma$); the $1e0p0\pi$ channel alone does not strongly discriminate. The paper concludes that MicroBooNE data are inconsistent with an electron-like interpretation of the MiniBooNE low-energy excess.
Load-bearing premise
The load-bearing premise is that an electron-like MiniBooNE excess would appear in MicroBooNE as charged-current $\nu_e$ interactions without visible pions in the two selected samples, with hadronic kinematics left unchanged from the standard simulation, and that systematic uncertainty on the signal models can be neglected.
Editorial extensions
If this is right
- The energy-dependent electron-neutrino enhancement (LEE Signal Model 1) is excluded at $>99\%$ CL$_s$ when both signal channels are combined in reconstructed neutrino energy.
- The new shower-energy/angle model (LEE Signal Model 2) is likewise excluded at $>99\%$ CL$_s$ in both reconstructed shower energy and shower $\cos\theta$.
- The $1e\mathrm{N}p0\pi$ channel provides most of the rejecting power; the observed 24% deficit in that channel (2.4$\sigma$) is itself unexplained and motivates improved $\nu_e$ cross-section models on argon.
- In the $1e0p0\pi$ channel alone the data are compatible with both the null hypothesis and a MiniBooNE-sized signal (best-fit signal strength up to 0.61), so the exclusion is not driven by that sample.
- Since the two-hypothesis tests are limited by statistics, future reconstruction and analysis improvements could sharpen the exclusion or reveal a residual excess in the same final states.
Reading between the lines
- The exclusion is conditional on the benchmark models: an electron-like excess that produces different hadronic final states, or that modifies hadronic kinematics rather than only electron kinematics, would not be covered by this test, so the MiniBooNE anomaly remains open.
- Because the paper neglects systematic uncertainty on the MiniBooNE LEE signal prediction, the $>99\%$ CL$_s$ numbers should be read as conditional; including those uncertainties would likely soften the confidence levels, though probably not erase the rejection.
- The 2.4$\sigma$ deficit in $1e\mathrm{N}p0\pi$ suggests the simulation overpredicts proton-tagged $\nu_e$ events; a corrected cross-section or hadronization model that removes this deficit would shift the background and could alter the CL$_s$ values in either direction.
- The result complements sterile-neutrino searches by tightening constraints on $\nu_e$ appearance interpretations of the LEE, while leaving $\nu_e$ disappearance and photon-based explanations untouched.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. This Letter analyzes the full five-year MicroBooNE dataset (1.11e21 POT) to search for excess charged-current electron-neutrino interactions without visible pions, motivated by the MiniBooNE low-energy excess. Two event selections are used, 1eNp0pi and 1e0p0pi, and the data are compared to the Standard Model prediction and to two empirical signal models derived from the MiniBooNE excess: Model 1 unfolds the excess in neutrino energy, and Model 2 unfolds it in shower energy and angle. The paper reports null-hypothesis p-values above 26.7% for the combined channels in all three kinematic variables, and combined CL_s exclusions of the two signal models at the 0.41%, 0.02%, and 0.014% level (1-CL_s), corresponding to >99% CL_s. The authors note that the 1e0p0pi channel alone yields much weaker exclusions and that the combined result is driven by 1eNp0pi.
Significance. If the benchmark exclusions are taken at face value, this is the strongest MicroBooNE statement to date against electron-like interpretations of the MiniBooNE LEE, using a 70% larger dataset than previous results, an expanded set of constraint channels, improved detector-systematics correlations, and a new two-dimensional signal model in shower energy and angle. The analysis is carefully executed: it uses sideband constraints, pseudo-data trials, the CL_s method, and Feldman-Cousins intervals, and it transparently reports the channel-by-channel results. The main value of the paper is the breadth of kinematic variables tested and the explicit construction of benchmark models that connect to MiniBooNE's observables. The reach of the conclusion, however, is bounded by assumptions in the signal-model construction that the paper itself only partially qualifies.
major comments (3)
- [Signal model] The signal prediction for both LEE models is obtained by scaling the true electron kinematics of MicroBooNE's intrinsic nu_e simulation while "leaving the modeling of the hadronic kinematics unchanged." This fixes the relative populations of the 1eNp0pi and 1e0p0pi channels to the Standard Model CC nu_e-on-argon prediction. Table I shows that the 1e0p0pi channel alone excludes Model 1 at 1-CL_s = 43% (neutrino energy) and Model 2 at 23% (shower energy) and 6.7% (shower cos theta), while the combined values are 0.41%, 0.02%, and 0.014%. The combined >99% CL_s exclusion is therefore driven almost entirely by the 1eNp0pi channel. If an electron-like LEE had a different hadronic topology (for example, no visible proton), the signal would preferentially populate 1e0p0pi, where the data are compatible with both H0 and H1, and the headline exclusion would not hold. The abstract and conclusions should either restrict the claim to the specific hadronic-topology assumption of the models or present a quantitative study of the robustness of the exclusion to variations in the proton-multiplicity split.
- [Signal model / Results] The paper states that the LEE signal models "for simplicity neglect systematic uncertainty on the MiniBooNE LEE signal prediction." Since the central claim is an exclusion of these models at >99% CL_s, the H1 predictions used in the CL_s and signal-strength tests do not include uncertainties from the unfolding of the MiniBooNE excess or from the model-building procedure. Depending on the size of these uncertainties, the reported CL_s values could be overconfident. The authors should either propagate this uncertainty into the CL_s calculation or provide a quantitative argument, for example by varying the unfolded MiniBooNE excess within its uncertainties, that the effect on the reported exclusions is negligible.
- [Results] The combined exclusion is driven by a channel in which the null prediction is itself in tension with data: the paper reports 102 observed events versus 133.5 +/- 7.4 predicted in 1eNp0pi, a 24% deficit with 2.4 sigma significance. The CL_s method protects against downward statistical fluctuations, but it does not protect against an upward bias in the H0 prediction. Because the sideband constraint increased the signal-channel predictions due to the underpredictions in the control samples, the robustness of the >99% CL_s exclusion depends on the absolute normalization of the constrained H0 prediction in 1eNp0pi. The paper should discuss what would happen to the exclusion if the null prediction in this channel were lower, for example by quoting a CL_s value with the constrained prediction renormalized to the data in a control region or by giving a sensitivity scan over the normalization of the 1eNp0pi prediction.
minor comments (5)
- [Abstract] There is a spacing typo: "withp-values" should be "with p-values".
- [Table I] The row labels in Table I, particularly "obs.H 0 −H 1 ∆χ2" and "1 - CL s [%]", are difficult to parse; the caption should define the test statistics and the CL_s quantity more explicitly.
- [Results] The paper notes that the 1D projections used for Model 2 are not independent because the model is defined in two dimensions, but the discussion is brief; a sentence on the expected correlation between the shower-energy and shower-angle p-values would help readers interpret the two exclusions as related rather than independent measurements.
- [Systematic uncertainties] The smoothing of detector-variation histograms with a pseudo-gaussian filter is described only briefly; since this can affect the bin-to-bin correlations and hence the sideband constraint, a one-sentence description of the smoothing width and its validation would improve reproducibility.
- [Signal model] The statement that the LEE signal is treated "in the same way as the electron-neutrinos selected" for the sideband constraint deserves an explicit caveat: the constraint is derived from nu_mu and pi0 control samples, so it inherits the assumption that the LEE has the same response to those constraints as intrinsic CC nu_e events.
Circularity Check
No significant circularity; external MiniBooNE templates and orthogonal control samples make the exclusion self-contained.
full rationale
The paper's central exclusion is not circular. The two LEE signal hypotheses are constructed from MiniBooNE's published excess via unfolding, which is an external benchmark independent of the MicroBooNE data being tested. The only parameter fitted to MicroBooNE data is the overall signal strength mu, and the headline claim is the CL_s rejection of mu=1 at the nominal MiniBooNE-sized signal; the fitted best-fit values are reported as fits, not repackaged as predictions. The null prediction in the signal channels is constrained using control samples (1muNp0pi, 1mu0p0pi, NCpi0) that are orthogonal to the 1eNp0pi and 1e0p0pi signal selections, so no signal bin is used to define its own expectation. The statement that the LEE signal model leaves 'the modeling of the hadronic kinematics unchanged' is a model assumption and a limitation on the scope of the exclusion, not a circular reduction. Self-citations to Ref. [4] carry over the selection, simulation, and systematic framework, but the current measurement, data, and statistical tests stand independently; no load-bearing claim is justified only by those citations. The paper therefore contains no step in which a predicted quantity equals, by construction, a fitted input or a self-cited result.
Assumptions & free parameters
free parameters (1)
- LEE signal strength mu =
Best fit 0.00 for combined channels in all variables; 2-sigma upper limits 0.22 to 0.50 depending on variable and…
assumptions (5)
- domain assumption The GENIE neutrino interaction model, flux simulation, particle transport, and detector response model accurately describe the standard-model backgrounds.
- ad hoc to paper The MiniBooNE low-energy excess can be fully represented as an electron-like CC nu_e signal whose unfolded kinematics from MiniBooNE data (model 1 in neutrino energy, model 2 in shower energy and angle) carry over to MicroBooNE.
- domain assumption The LEE signal events would pass the 1eNp0pi and 1e0p0pi selections and leave the hadronic final state unchanged relative to the intrinsic nu_e prediction.
- standard math Systematic uncertainties can be represented as multivariate normal variations with covariance obtained from event reweighting and detector model variations, and Poisson data fluctuations added as a Neyman-Pearson term.
- standard math The block matrix method [32] correctly propagates constraints from the muon and pi0 control samples to the signal predictions.
Cite this review
Pith. "Pith review of Search for an Anomalous Production of Charged-Current $\nu_e$ Interactions Without Visible Pions Across Multiple Kinematic Observables in MicroBooNE." pith.science (2026). https://pith.science/paper/72AYZPD5
@misc{pith2026241214407,
author = {Pith},
title = {Pith review of: Search for an Anomalous Production of Charged-Current $\nu_e$ Interactions Without Visible Pions Across Multiple Kinematic Observables in MicroBooNE},
year = {2026},
howpublished = {\url{https://pith.science/paper/72AYZPD5}},
note = {Machine review of arXiv:2412.14407}
}
abstract
This Letter presents an investigation of low-energy electron-neutrino interactions in the Fermilab Booster Neutrino Beam by the MicroBooNE experiment, motivated by the excess of electron-neutrino-like events observed by the MiniBooNE experiment. This is the first measurement to use data from all five years of operation of the MicroBooNE experiment, corresponding to an exposure of $1.11\times 10^{21}$ protons on target, a $70\%$ increase on past results. Two samples of electron neutrino interactions without visible pions are used, one with visible protons and one without any visible protons. The MicroBooNE data show reasonable agreement with the nominal prediction, with $p$-values $\ge 26.7\%$ when the two $\nu_e$ samples are combined, though the prediction exceeds the data in limited regions of phase space. The data is further compared to two empirical models that modify the predicted rate of electron-neutrino interactions in different variables in the simulation to match the unfolded MiniBooNE low energy excess. In the first model, this unfolding is performed as a function of electron neutrino energy, while the second model aims to match the observed shower energy and angle distributions of the MiniBooNE excess. This measurement excludes an electron-like interpretation of the MiniBooNE excess based on these models at $> 99\%$ CL$_\mathrm{s}$ in all kinematic variables.
Figures
Forward citations
Cited by 1 Pith paper
-
The High W Challenge: Robust Neutrino Energy Estimators for LArTPCs
The W²-based estimator shows the smallest bias versus true neutrino energy and greater stability to mismodelling of scattering and interactions than four common alternatives in LArTPC experiments.
Reference graph
Works this paper leans on
-
[4]
Abratenkoet al.(MicroBooNE Collaboration), Phys
P. Abratenkoet al.(MicroBooNE Collaboration), Phys. Rev. D105, 112004 (2022)
work page 2022
-
[1]
A. A. Aguilar-Arevaloet al.(MiniBooNE Collaboration), Phys. Rev. Lett.121, 221801 (2018)
work page 2018
-
[2]
Aguilaret al.(LSND Collaboration), Phys
A. Aguilaret al.(LSND Collaboration), Phys. Rev. D 64, 112007 (2001)
work page 2001
-
[3]
Acciarriet al.(MicroBooNE Collaboration), J
R. Acciarriet al.(MicroBooNE Collaboration), J. In- strum.12, P02017 (2017)
work page 2017
-
[5]
Abratenkoet al.(MicroBooNE Collaboration), Phys
P. Abratenkoet al.(MicroBooNE Collaboration), Phys. Rev. Lett.128, 241801 (2022)
work page 2022
-
[6]
Abratenkoet al.(MicroBooNE Collaboration), Phys
P. Abratenkoet al.(MicroBooNE Collaboration), Phys. Rev. D105, 112003 (2022)
work page 2022
-
[7]
Abratenkoet al.(MicroBooNE Collaboration), Phys
P. Abratenkoet al.(MicroBooNE Collaboration), Phys. Rev. D105, 112005 (2022)
work page 2022
-
[8]
Abratenkoet al.(MicroBooNE Collaboration), Phys
P. Abratenkoet al.(MicroBooNE Collaboration), Phys. Rev. Lett.128, 111801 (2022)
work page 2022
Show all 41 references
-
[9]
A. A. Aguilar-Arevaloet al.(MiniBooNE Collaboration), Phys. Rev. D103, 052002 (2021)
2021
-
[10]
K. J. Kelly and J. Kopp, JHEP2023, 113 (2023)
2023
-
[11]
Brdar and J
V. Brdar and J. Kopp, Phys. Rev. D105, 115024 (2022)
2022
-
[12]
A. Diaz, C. Arg¨ uelles, G. Collin, J. Conrad, and M. Shae- vitz, Phys. Rep.884, 1 (2020)
2020
-
[13]
B¨ oser, C
S. B¨ oser, C. Buck, C. Giunti, J. Lesgourgues, L. Lud- hova, S. Mertens, A. Schukraft, and M. Wurm, Prog. Part. Nucl. Phys.111, 103736 (2020)
2020
-
[14]
Vergani, N
S. Vergani, N. W. Kamp, A. Diaz, C. A. Arg¨ uelles, J. M. Conrad, M. H. Shaevitz, and M. A. Uchida, Phys. Rev. D104, 095005 (2021)
2021
-
[15]
Alvarez-Ruso and E
L. Alvarez-Ruso and E. Saul-Sala, inProspects in Neu- trino Physics(2017) arXiv:1705.00353 [hep-ph]
2017 arXiv
-
[16]
Fischer, A
O. Fischer, A. Hern´ andez-Cabezudo, and T. Schwetz, 8 Phys. Rev. D101, 075045 (2020)
2020
-
[17]
Abdullahi, M
A. Abdullahi, M. Hostert, and S. Pascoli, Phys. Lett. B 820, 136531 (2021)
2021
-
[18]
Bertuzzo, S
E. Bertuzzo, S. Jana, P. A. N. Machado, and R. Zukanovich Funchal, Phys. Rev. Lett.121, 241801 (2018)
2018
-
[19]
Ballett, S
P. Ballett, S. Pascoli, and M. Ross-Lonergan, Phys. Rev. D99, 071701 (2019)
2019
-
[20]
Abratenkoet al.(MicroBooNE Collaboration), Phys
P. Abratenkoet al.(MicroBooNE Collaboration), Phys. Rev. D105, 072001 (2022)
2022
-
[21]
Allisonet al., Nucl
J. Allisonet al., Nucl. Instrum. Meth. A835, 186 (2016)
2016
-
[22]
Allisonet al., IEEE Trans
J. Allisonet al., IEEE Trans. Nucl. Sci.53, 270 (2006)
2006
-
[23]
Agostinelliet al., Nucl
S. Agostinelliet al., Nucl. Instrum. Meth. A506, 250 (2003)
2003
-
[24]
Acciarriet al.(MicroBooNE Collaboration), Eur
R. Acciarriet al.(MicroBooNE Collaboration), Eur. Phys. J. C78, 82 (2018)
2018
-
[25]
Adamset al.(MicroBooNE Collaboration), J
C. Adamset al.(MicroBooNE Collaboration), J. In- strum.14, P04004 (2019)
2019
-
[26]
Adamset al.(MicroBooNE Collaboration), J
C. Adamset al.(MicroBooNE Collaboration), J. In- strum.15, P02007 (2020)
2020
-
[27]
M. J. Berger, J. S. Coursey, M. A. Zucker, and J. Chang, Stopping-Power & Range Tables for Electrons, Protons, and Helium Ions, NIST Standard Reference Database No. 124 (National Institute of Standards and Technol- ogy, 2017)
2017
-
[28]
D. E. Groom, N. V. Mokhov, and S. I. Striganov, Atomic Data and Nuclear Data Tables78, 183 (2001)
2001
-
[29]
Abratenkoet al.(MicroBooNE), JHEP12, 153 (2021)
P. Abratenkoet al.(MicroBooNE), JHEP12, 153 (2021)
2021
-
[30]
Abratenkoet al.(MicroBooNE Collaboration), Eur
P. Abratenkoet al.(MicroBooNE Collaboration), Eur. Phys. J. C82, 454 (2022)
2022
-
[31]
See Supplemental Material, which includes Refs. [4, 25, 41], for additional information for this analysis, including the impact of the CRT on selection, constraint covariance and its impact, smoothing of detector systematics, data validation, updates to the signal model, a sum...
-
[32]
M. L. Eaton,Multivariate statistics, Probability & Math- ematical Statistics (John Wiley & Sons, Nashville, TN,
-
[33]
X. Ji, W. Gu, X. Qian, H. Wei, and C. Zhang, Nucl. Instrum. Meth. A961, 163677 (2020)
2020
-
[34]
A. A. Aguilar-Arevaloet al.(MiniBooNE Collaboration), Phys. Rev. Lett.98, 231801 (2007)
2007
-
[35]
Abratenkoet al.(MicroBooNE Collaboration), Phys
P. Abratenkoet al.(MicroBooNE Collaboration), Phys. Rev. D106, L051102 (2022)
2022
-
[36]
G. J. Feldman and R. D. Cousins, Phys. Rev. D57, 3873 (1998)
1998
-
[37]
Abratenkoet al.(MicroBooNE Collaboration), Phys
P. Abratenkoet al.(MicroBooNE Collaboration), Phys. Rev. D105, L051102 (2022)
2022
-
[38]
Abratenkoet al.(MicroBooNE Collaboration), Phys
P. Abratenkoet al.(MicroBooNE Collaboration), Phys. Rev. Lett.130, 011801 (2023)
2023
-
[39]
Junk, Nucl
T. Junk, Nucl. Instrum. Meth. A434, 435 (1999)
1999
-
[40]
A. L. Read 10.5170/CERN-2000-005.81 (2000)
2000 doi
-
[41]
MicroBooNE Collaboration, MICROBOONE-NOTE- 1043-PUB, DOI:10.2172/1573217 (2018)
2018 doi
Reviewed August 11, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.