Skip to content
CosmosExplorer

Modern Slow-Roll Inflation

1983 / 2018 · Andrei Linde, Alexei Starobinsky, many others
Strongly supported

An effective field theory framework where the early universe is dominated by a scalar field whose flat potential drives quasi-exponential expansion, generating the perturbation spectrum measured by Planck.

Skip 3D content

In one sentence

Modern slow-roll inflation treats the inflationary epoch as an effective field theory of a scalar field on a flat potential, with many candidate potentials consistent with Planck's measured scalar spectral index and tensor-to-scalar ratio constraints.

The claim

Modern inflation is formulated as an effective field theory: a scalar field with potential V coupled to gravity, satisfying slow-roll conditions where the field's kinetic energy is small compared to its potential energy. The universe undergoes accelerated expansion for at least 50 to 60 e-folds, solving the horizon, flatness, and monopole problems.

Quantum fluctuations of the scalar field on sub-Hubble scales are stretched to super-Hubble scales by inflation, freezing in as classical curvature perturbations. The framework is flexible: many specific potentials (chaotic, plateau, hilltop, Starobinsky) are consistent with data, and slow-roll formulas give universal predictions for the scalar spectral index n_s, tensor-to-scalar ratio r, and consistency relations.

The family stance

Our universe began with a brief epoch of exponential expansion driven by a scalar field, followed by reheating into the hot Big Bang phase. The same inflaton field that drove expansion also generated the seed perturbations that became galaxies.

Predictions

  • Scalar spectral index n_s approximately 0.965, slightly less than 1, consistent across many viable potentials
  • Gaussian, adiabatic perturbations with minimal non-Gaussianity (small f_NL)
  • Tensor-to-scalar ratio r constrained by potential shape, with the consistency relation n_t = -r/8 for single-field slow-roll
  • Specific acoustic peak structure and polarization patterns in the CMB

Evidence

  • Planck 2018 measures n_s = 0.965, excluding exact scale invariance at many sigma and matching slow-roll predictions.
  • Combined Planck and BAO data constrain spatial curvature to near zero, consistent with inflation's flatness prediction.
  • BICEP/Keck and Planck place upper bounds r < 0.03, ruling out simple large-field models but consistent with many viable slow-roll potentials.
  • The detailed acoustic peak structure of the CMB reflects coherent initial phases consistent with inflation-generated perturbations.

Counterpoints

  • Initial conditions for inflation itself may require a fine-tuned homogeneous patch and specific scalar field values.
  • Embedding viable inflaton potentials in a UV-complete quantum gravity (e.g., string theory) remains nontrivial; swampland conjectures challenge whether sufficiently flat potentials are generic.
  • Most slow-roll potentials lead to eternal inflation in some regime, raising measure and predictability concerns.
0votes
Go deeperTechnical detail with proper terminology

The slow-roll parameters epsilon and eta are combinations of V'/V and V''/V; specific potential shapes determine n_s and r via universal formulas. See Lyth and Riotto (1999) for the standard reference.

Planck 2018 results constrain over a dozen specific inflationary potentials, ruling some out (e.g., simple phi^4 chaotic) while leaving plateau models like Starobinsky inflation as best fits.

References

  1. Established
    Linde (1983) Chaotic inflation, Phys. Lett. B 129, 177
  2. Established
    Starobinsky (1980) A new type of isotropic cosmological models without singularity, Phys. Lett. B 91, 99
  3. Established
    Lyth & Riotto (1999) Particle physics models of inflation and the cosmological density perturbation, Phys. Rept. 314, 12,307 citations
  4. Established
    Planck Collaboration (2020) Planck 2018 results X. Constraints on inflation, Astron. Astrophys. 641, A10

Last reviewed May 15, 2026

Spotted an error? Have a source to add?

Prefer email?

You can also send a prefilled email with the variant URL already filled in.

Related theories