Another Search For Lepton Universality Violation Comes Up Empty

While there are tensions with the Standard Model prediction of lepton universality (i.e. that charged leptons have the same properties except mass), from a couple of kinds of B meson decay, only one of which exceeds three sigma, a test of isospin partners of those same decays (albeit with less precision) find no statistically significant deviation from the Standard Model prediction, making the anomaly less likely to be real.

Tests of lepton universality in B0→K0Sl+l− and B+→K∗+ℓ+ℓ− decays where l is either an electron or a muon are presented. The differential branching fractions of B0→K0Sl+l− and B+→K∗+ℓ+ℓ− decays are measured in intervals of the dilepton invariant mass squared. The measurements are performed using proton-proton collision data recorded by the LHCb experiment, corresponding to an integrated luminosity of 9fb−1. The results are consistent with the Standard Model and previous tests of lepton universality in related decay modes. The first observation of B0→K0Sl+l− and B+→K∗+ℓ+ℓ− decays is reported.

LHCb collaboration, "Tests of lepton universality using B0→K0Sl+l− and B+→K∗+ℓ+ℓ− decays" arXiv:2110.09501 (October 18, 2021).

This tends to support an interpretation of the prior lepton universality violation indications as a fluke or some sort of unrecognized systemic error because these are isospin partners of the cases where apparent violations were seen. The body text of the Letter explains this while also noting that the reduced precision of this measurement is also a factor:

Forces in the SM couple to the charged leptons with equal strength, which is referred to as lepton universality. Therefore, these ratios are predicted to be very close to unity, with small corrections due to the muon-electron mass difference. Furthermore, these ratios benefit from precise cancellation of the hadronic uncertainties that affect predictions of the branching fractions and angular observables. Significant deviation from unity in such ratios would therefore constitute unambiguous evidence of BSM physics.

The ratio RK∗0 , measured by the LHCb collaboration using the data collected in the q 2 regions 0.045 < q2 < 1.1 GeV2 /c4 and 1.1 < q2 < 6.0 GeV2 /c4 , is in tension with the SM predictions at 2.2–2.4 and 2.4–2.5 standard deviations (σ), respectively. A measurement of RK+ performed in the region 1.1 < q2 < 6.0 GeV2 /c4 deviates from the SM by 3.1 standard deviations. The analogous ratio measured using Λ 0 b → pK−` +` − decays, RpK, is consistent with the SM within one standard deviation. All four measurements show a deficit of b→ sµ+µ − decays with respect to b→ se+e − decays.

In addition, angular observables and branching fractions of b→ sµ+µ− decays have been measured, with several in tension with the SM. However, the extent to which they may be affected by residual quantum chromodynamics contributions remains uncertain. Intriguingly, it is possible to account for all these anomalies simultaneously through the modification of the b→ s coupling in a model-independent way. Such a modification can be generated by the presence of a heavy neutral boson or a leptoquark, as well as in models with supersymmetry, extra dimensions, and extended Higgs sectors.

The B0→ K0 S ` +` − and B+→ K∗+` +` − decays are the isospin partners of B+→ K+` +` − and B0→ K∗0 ` +` − decays and are expected to be affected by the same NP contributions. Testing lepton universality by measuring the ratios RK0 S and RK∗+ can therefore provide important additional evidence for or against NP. However, while these decays have similar branching fractions to their isospin partners, O(10−6 ) to O(10−7 ), they suffer from a reduced experimental efficiency at LHCb due to the presence of a long-lived K0 S or π 0 meson in the final state.

Proponents of the anomaly argue that the central values weren't that much different from previous ones despite a lack of statistical significance. 

It is also worth considering that a b-->s transition does not happen at tree level in the Standard Model. It requires a W- boson interaction from a b quark to an up-like quark followed by a transition from an up-like quark to an s quark in a W+ interaction at the one-loop level, or a further iteration of up-quark and down-quark pin pong at the three and/or five loop level. 

And, an sµ+µ− end product can decay to an se+e- end product plus a pair of neutrinos via an additional round of weak force interactions by the muons (along with additional decay products, some of which would cancel out virtually or be to invisible neutrinos), although these would usually be rather slow compared to the other decays (particularly if the muons are traveling at the relativistic speeds that they would be with these kinds of energies).

The complexity of the possible paths makes it harder to model correctly.