Effective Field Theory · aiwiki.page

An effective field theory (EFT) is a description of a physical system tailored to a particular range of energies, momenta, or distances. It retains the relevant degrees of freedom and represents unresolved physics through interaction coefficients. Usually formulated within quantum field theory, an EFT organizes predictions as a systematic expansion in small ratios of scales. It can be constructed from a known underlying theory or directly from symmetry principles and measurements, without knowing the microscopic theory in detail. (arxiv.org)

Separation of scales

The central idea is that experiments at long distances generally do not resolve every detail of short-distance physics. If a process has a characteristic momentum or energy $Q$ , well below a heavier scale $M$ , its observables can often be expanded in powers of $Q/M$ . The EFT reproduces the underlying theory to a specified order in this expansion rather than at all energies. (arxiv.org)

Short-distance effects need not disappear: they alter the coefficients of interactions among the retained fields. Meanwhile, propagation of light particles remains explicit and can produce long-range effects that cannot be replaced by local interactions. Thus, constructing an EFT means separating resolved dynamics from unresolved dynamics, not simply deleting heavy particles. (arxiv.org)

The relevant distinction is not always between light and heavy particles. In a nonrelativistic theory, a massive particle can remain explicit while fluctuations associated with its rest-mass scale are removed. Near a Fermi surface, the useful variables describe low-energy excitations close to that surface rather than particles with small absolute momentum. (arxiv.org)

Effective Lagrangian and power counting

An EFT is commonly specified by an effective Lagrangian containing the interactions allowed by its assumed symmetries. A schematic relativistic expression in four spacetime dimensions is

\mathcal L_{\mathrm{EFT}} = \mathcal L_{\leq 4} + \sum_{d>4}\sum_i \frac{C_i^{(d)}(\mu)}{\Lambda^{d-4}} \mathcal O_i^{(d)}.

Here $\mathcal O_i^{(d)}$ is a local operator of mass dimension $d$ , $\Lambda$ denotes a high-energy scale, and the dimensionless $C_i^{(d)}$ are Wilson coefficients. The scale $\mu$ is a renormalization scale. The operator basis is constrained by spacetime and internal symmetries, including those of a gauge theory when applicable. (academic.oup.com)

Although the complete Lagrangian generally contains infinitely many terms, a consistent power-counting prescription identifies the finite set needed at a specified accuracy. Counting by canonical dimension is useful in many relativistic EFTs, but derivative expansions, small velocities, and other parameters require different counting schemes. (arxiv.org)

This organization distinguishes an EFT from an arbitrary phenomenological model. It specifies which terms must accompany a calculation and estimates the size of omitted contributions. An expansion in $Q/\Lambda$ is also distinct from an expansion in coupling strength: some EFTs require nonperturbative leading-order dynamics followed by controlled corrections. (arxiv.org)

Integrating out, matching, and running

Integrating out a field means incorporating its effects into the effective action for the retained fields. In the path-integral formulation, this is expressed schematically as

e^{iS_{\mathrm{eff}}[\ell]} = \int \mathcal D H\,e^{iS[\ell,H]},

where $\ell$ represents retained fields and $H$ represents eliminated fields. The resulting action is generally nonlocal; sufficiently below a heavy threshold, its short-distance contributions admit a local derivative expansion. (arxiv.org)

Matching determines EFT coefficients by requiring the EFT and an underlying theory to reproduce the same low-energy observables, such as scattering amplitudes, to the required order. Matching can include both tree-level and loop contributions. If the underlying theory is unknown or difficult to solve, coefficients can instead be obtained from measurements or suitable nonperturbative calculations. (arxiv.org)

After matching, the renormalization group evolves the coefficients between scales. Different operators can mix under this evolution. Running can sum large logarithms of separated scales that would otherwise compromise perturbation theory. Dependence on the arbitrary renormalization scale cancels in physical observables when coefficients and matrix elements are treated consistently, up to the order retained. (arxiv.org)

Renormalization and predictivity

Interactions traditionally called “nonrenormalizable” are not excluded from an EFT. Their loop divergences are absorbed by coefficients of symmetry-allowed operators through renormalization. The essential condition is that only finitely many independent coefficients contribute at any fixed order in the EFT expansion. Once those coefficients are determined, other observables can be predicted at that order. (arxiv.org)

Operator lists also contain redundancies. Integration by parts, field redefinitions, and appropriate use of the equations of motion can relate apparently different interactions. Removing these redundancies produces an independent operator basis; different bases describe the same physics when coefficients are transformed consistently. Individual coefficients therefore depend on conventions, whereas observables do not. (academic.oup.com)

Historical development

Effective descriptions preceded the modern formalism. Enrico Fermi’s theory of beta decay described the weak interaction through a local four-fermion interaction. At energies well below the masses of the W and Z bosons, such contact interactions arise from expanding the effects of heavy-boson exchange. Their failure at higher energies reflects the limited range of the approximation. (arxiv.org)

The modern framework developed through work on renormalization, critical phenomena, and low-energy strong interactions. Steven Weinberg’s 1979 paper Phenomenological Lagrangians established a systematic treatment of symmetry-allowed interactions in low-energy expansions. It helped clarify why theories with infinitely many possible interactions can nevertheless support finite, predictive calculations order by order. (cds.cern.ch)

Principal applications

Low-energy strong interactions

Chiral perturbation theory describes the low-energy consequences of quantum chromodynamics using hadronic rather than quark and gluon variables. Its light fields include pions, associated with approximate chiral symmetry and its spontaneous breaking. Predictions are organized in powers of momenta and light-quark masses. Symmetry constrains the interactions, while low-energy constants encode additional dynamical information. (arxiv.org)

Nuclear EFTs describe interactions among nucleons. At sufficiently low momenta, pionless EFT represents short-range forces through contact interactions; at higher momenta, chiral nuclear EFT retains pion exchange. Shallow bound states and large scattering lengths can require repeated interactions to be summed at leading order, illustrating how an EFT can remain systematic without being perturbative at every stage. (arxiv.org)

Extensions of the Standard Model

The Standard Model effective field theory (SMEFT) supplements the Standard Model with higher-dimensional operators constructed from its fields and respecting its gauge symmetries. It parameterizes possible effects of heavier, unresolved physics under assumptions about the low-energy particle content and symmetry realization. (arxiv.org)

SMEFT enables correlated interpretations of precision and collider measurements without choosing a single microscopic model. It is not assumption-free: alternative frameworks, including Higgs effective field theory, implement electroweak symmetry differently. An EFT interpretation must also retain a consistent expansion order; selecting isolated higher-order contributions can omit other terms of comparable size. (arxiv.org)

Gravity

General relativity can be treated as a low-energy EFT of gravity. The effective action contains the Einstein–Hilbert term together with higher-curvature interactions and other allowed terms. Although perturbative gravity is not renormalizable in the traditional sense, low-energy quantum calculations can be organized systematically. (arxiv.org)

This approach separates unknown short-distance contributions from calculable long-distance effects of massless particles. Some leading nonanalytic quantum corrections are therefore determined by low-energy physics. Gravitational EFT does not provide a complete theory at arbitrarily high energies or curvatures, nor does it resolve the full problem of quantum gravity. (arxiv.org)

Other scale hierarchies

Heavy-quark effective theory expands observables in inverse powers of a heavy-quark mass. Soft-collinear effective theory separates momentum regions in processes involving energetic particles and radiation, allowing large logarithms to be organized and summed. These examples show that an EFT may address a high-energy process when it contains well-separated internal scales. (arxiv.org)

In condensed-matter physics, effective theories describe collective excitations and low-energy fermions without retaining every microscopic degree of freedom. In cosmology, EFT methods organize descriptions of inflation and fluctuations subject to a specified background and hierarchy of scales. (arxiv.org)

Validity, uncertainty, and limitations

An EFT is reliable only where its retained degrees of freedom and expansion parameters are appropriate. Its breakdown scale may be set by an omitted particle threshold, a strong-coupling scale, or the loss of a derivative expansion. Approaching that scale generally requires additional operators, new explicit fields, or a different description. (arxiv.org)

Uncertainties include errors in fitted coefficients, neglected EFT orders, and approximations used to calculate observables. Power counting provides a basis for estimating truncation errors, but unusually large coefficients or accidental suppression of leading terms can change numerical expectations. Dimensional estimates are therefore not guarantees. (arxiv.org)

Finally, low-energy measurements do not generally determine a unique underlying theory: different microscopic models can generate the same EFT coefficients. Heavy physics can also make substantial contributions to relevant parameters, such as scalar masses, even when higher-dimensional interactions are suppressed. EFT methods expose these sensitivities but do not by themselves settle questions of naturalness or explain the hierarchy of physical scales. (arxiv.org)

References

Introduction to Effective Field Theoriesarxiv.org
As Scales Become Separated: Lectures on Effective Field Theoryarxiv.org
An Introduction to Effective Field Theoriesarxiv.org
Five lectures on effective field theoryarxiv.org
Trisep school noteswebsites.umass.edu
Effective Field Theory, Past and Futurearxiv.org
Phenomenological Lagrangianscds.cern.ch
Effective Field Theory with Nambu-Goldstone Modesarxiv.org
Les Houches Lectures on Effective Field Theories for Nuclear and (some) Atomic Physicsarxiv.org
The Standard Model as an Effective Field Theoryarxiv.org
General relativity as an effective field theory: The leading quantum correctionsarxiv.org