If vector-like leptons modify both loops and tree decays, how do global fits disentangle their individual contributions?

Question

Antigravity · Accepted Answer

This is a sophisticated question about how global fits separate tree level and loop level contributions from a single class of new particle — vector like leptons. Let me work through the concrete strategy. The Disentanglement Challenge Vector like leptons contribute to observables through two distinct mechanisms simultaneously [AG 2026.03 1618]: they modify decay rates at tree level (direct quark lepton couplings) and generate loop corrections that alter precision observables like magnetic moments and rare decays. A naive fit to one observable at a time cannot tell you whether a deviation comes from the tree or loop channel — or both in combination. How Global Fits Separate Them The key insight is operator mixing under renormalization group evolution [AG 2024.08 976]. When you evolve the effective field theory Hamiltonian from high scales (where new physics enters) down to low scales, tree level operators generate loop suppressed operator contributions of different types. This means: 1. Different observables probe different operator combinations. For instance, leptonic $B$ decays are sensitive to direct quark lepton couplings [AG 2025.11 1220], while muon g−2 picks up dipole moment contributions [AG 2024.08 1174]. They have different momentum dependences and helicity structures. 2. Precision measurements constrain the parameter space in complementary ways. A global fit simultaneously fits branching fractions (sensitive to tree amplitudes), dipole moments (loop induced), and collider constraints (production thresholds and decay widths). The vector like lepton mass, mixing angles, and coupling strengths are forced to satisfy all of these at once [AG 2026.03 1618]. 3. Loop effects can dominate bounds even when tree contributions are small. One loop corrections sometimes yield stronger constraints on new physics scales than tree level processes [AG 2025.03 1775]. A global fit automatically weights these appropriately. 4. Identifying flat directions eliminates degeneracies. When multiple operator combinations affect the same observable identically, fits have "flat directions" — directions in parameter space along which the likelihood doesn't change. Global analyses across lepton flavor conserving and violating channels, plus collider observables, typically break these degeneracies [AG 2024.03 1124]. Concrete Example In the $U 1$ vector leptoquark model [AG 2025.11 1220], the branching fraction of $B^+ 	o \ell^+ 
u$ gets contributions from both the direct tree level amplitude and modified loop level $B$ meson decay constants. A fit to this decay alone cannot separate them. But when you simultaneously fit to: $B$ to dileptons (tree and loop sensitive in different ratios) Lepton dipole moments (loop dominated) [AG 2024.08 1174] $Z$ and Higgs decays (constrain couplings independently) [AG 2026.03 1618] ...the redundancy of constraints forces the tree and loop contributions into definite ratios dictated by the underlying model structure. Why It Matters This disentanglement is crucial because tree and loop contributions often have opposite signs or different sensitivity to coupling strengths. A global fit tells you not just "there's new physics" but how it enters—whether mainly as direct interactions or virtual particle effects—which guides experimental searches and future collider design. The provided preprints give excellent case studies but don't explicitly describe the abstract methodology of global fitting itself; you'd want a dedicated review on effective field theory global analysis techniques for a fuller technical account.