Three Advanced Lectures on Inflation

Abstract: Lecture notes on inflation. The lectures are three double lectures, held for the first time at the Nordita Winter School 2024 - Particle Physics and Cosmology, covering an advanced introduction to the theory of primordial inflation, as well as the linear and non-linear perturbation theory of slow-roll inflation/quasi-de Sitter spacetimes.

• The paper presents comprehensive lectures on inflationary cosmology, detailing the causal structure, microphysical models, and perturbation theory that underpin the early universe.
• It rigorously derives inflationary observables like the scalar power spectrum, spectral tilt, and tensor-to-scalar ratio using slow-roll formalism and consistency relations.
• The work also explores nonlinear effects, non-Gaussianity, and the links between asymptotic symmetries and infrared phenomena, setting a framework for future observational tests.

Advanced Overview of Inflationary Cosmology and Perturbation Theory

Motivation for Inflation and the Causality Problem

The standard Big Bang model suffers from fundamental shortcomings, especially the horizon (causality) and flatness problems. The observed homogeneity of the CMB across causally disconnected regions at the time of last scattering is inconsistent with the absence of inflationary dynamics. The Penrose diagrams elucidated in the lectures concisely summarize both the causal structure of different FLRW cosmologies and motivate the necessity of an early de Sitter-like (inflationary) phase. Specifically, in the absence of inflation, null geodesics corresponding to our past light cone at recombination do not overlap, which is incompatible with the isotropy of the CMB. Introducing a quasi-de Sitter epoch, with $a(t) \propto e^{Ht}$ and a vacuum energy equation of state $p=-\rho$, pushes the initial singularity arbitrarily far back and allows for the entire observable universe to have been in causal contact prior to the hot Big Bang phase.

The number of inflationary $e$-folds $N \gtrsim 60$ required to resolve the horizon and flatness problems is derived precisely, depending on the scale of reheating and the Hubble scale during inflation. The mechanism also exponentially dilutes any pre-existing spatial curvature, enforcing $\Omega \to 1$ as an attractor solution.

Microphysical Models of Inflation

Old and New Approaches

Early models such as "old inflation" relied on a scalar field trapped in a metastable minimum, with vacuum energy driving de Sitter expansion. First-order phase transitions ("bubble nucleation") fail due to the graceful exit problem: percolation in an expanding background is insufficient to reheat the universe globally and homogenously.

Slow-Roll Inflation

The contemporary paradigm is slow-roll dynamics for a scalar ("inflaton") minimally coupled to gravity, governed by a potential $V(\phi)$. Slow-roll parameters ($\epsilon$, $\eta$) quantify the smallness of kinetic and potential derivatives, ensuring $\epsilon, |\eta| \ll 1$, which allows an extended phase of quasi-de Sitter expansion with controlled departures from scale invariance. The slow-roll formalism relates inflationary observables to the inflationary potential.

Inflationary Model Classification

A systematic taxonomy is provided:

• Large-field models: $\Delta\phi \gg M_{p}$. The prototype is chaotic inflation with monomial potentials, which can be protected by underlying shift symmetries (e.g., axion monodromy).
• Small-field models: $p=-\rho$0; inflation occurs near the origin. Potentials are flatter near the top, motivated by spontaneous symmetry breaking scenarios (hilltop and new inflation).
• Hybrid models: A second (waterfall) field triggers the end, facilitating inflation with sub-Planckian field excursions. Realized typically in SUSY/SUGRA contexts.
• Curvaton mechanism: A light spectator field generates perturbations distinct from the inflaton, allowing the inflaton potential to be steeper and compatible with CMB observables even if the inflaton-generated perturbations are suppressed.

Linear Perturbation Theory and Inflationary Signatures

Cosmological Perturbations

The lectures derive the power spectrum for scalar (curvature) perturbations, $p=-\rho$1, in detail. A gauge-invariant treatment exploiting the comoving gauge and ADM formalism leads to the quadratic action for $p=-\rho$2, with canonical quantization naturally leading to nearly scale-invariant, Gaussian fluctuations. The power spectrum is

$p=-\rho$3

where $p=-\rho$4 and $p=-\rho$5 are evaluated at horizon crossing.

The spectral tilt is a core prediction,

$p=-\rho$6

which is observed to be small and negative, in world-class agreement with Planck constraints ($p=-\rho$7) for potentials yielding $p=-\rho$8.

Tensor Modes and Consistency Relations

The lectures rigorously derive the tensor, or primordial gravitational wave, spectrum,

$p=-\rho$9

which sets up the key observable $e$0 (tensor-to-scalar ratio), linking $e$1 directly to the slope of the potential and the amplitude of gravitational waves. The single-field slow-roll "consistency relation" $e$2 persists as a sharp discriminant for inflationary model-building.

Curvaton and Mixed Scenarios

Detailed attention is given to scenarios where the primordial curvature perturbation is sourced not by the inflaton but by a light spectator (curvaton or modulus). The spectrum's tilt and amplitude are calculable, and such models often predict larger non-Gaussian signals.

Nonlinear Perturbation Theory, Non-Gaussianity, and Consistency Relations

Bispectrum and Trispectrum

The lectures systematically construct higher-point correlators, focusing on the bispectrum $e$3 and trispectrum $e$4. For single-field slow-roll inflation, the local shape non-Gaussianity is suppressed, as encapsulated by the Maldacena consistency condition,

$e$5

This prediction is robust: single-field attractor models with Bunch-Davies initial state yield unobservably small local $e$6. In contrast, curvaton scenarios generically lead to $e$7 when the curvaton dominates at decay, testable with future CMB surveys.

SSV Consistency Relation, Soft Limits, and Infrared Triangle

The advanced treatment extends the analysis to the four-point function (trispectrum), deriving the Seery-Sloth-Vernizzi (SSV) consistency relation for "collapsed" configurations in terms of the power spectrum and $e$8. These are direct consequences of large diffeomorphisms and symmetry properties of de Sitter space.

The lectures cover innovative aspects related to loop corrections and IR divergences in inflationary correlators (Giddings-Sloth relations), and the deep connection between cosmological soft theorems (Maldacena, SSV, GS) and asymptotic symmetries (de Sitter BMS analogs). A crucial conceptual framework is the "infrared triangle" linking asymptotic symmetries, soft theorems, and memory effects.

Implications for Cosmology and Prospects

Observational Constraints and Model Differentiation

Precision Planck data tightly constrains $e$9, with Starobinsky $N \gtrsim 60$0 inflation lying at the best-fit point in the $N \gtrsim 60$1-$N \gtrsim 60$2 plane, while simple monomial large-field models (e.g., $N \gtrsim 60$3) are highly disfavored. The non-detection of primordial gravitational waves ($N \gtrsim 60$4 at 95% CL) and the measurement of $N \gtrsim 60$5 to be consistent with zero are in sharp numerical agreement with slow-roll predictions.

However, the persistent Hubble tension motivates early dark energy scenarios, modifying pre-recombination expansion and favoring models with $N \gtrsim 60$6—in tension with canonical Starobinsky-like predictions and instead compatible with curvaton or mixed models (2606.06581).

Future Directions

Upcoming CMB and LSS experiments will probe tensor-to-scalar ratios down to $N \gtrsim 60$7 and $N \gtrsim 60$8 at the $N \gtrsim 60$9 level, providing critical discriminants among the various inflationary paradigms. Enhanced precision in primordial correlator measurements will allow detailed mapping of allowed parameter space, pinning down the inflaton potential and high-energy physics signatures.

On the theoretical front, continued development of consistency relations, summation and resummation of IR effects, and connections between asymptotic symmetries and cosmological observables offer promising routes to a more rigorous quantum gravity embedding of inflation.

Conclusion

The lectures in "Three Advanced Lectures on Inflation" (2606.06581) provide a comprehensive formal and technical introduction to modern inflationary cosmology, emphasizing the central role of causal structure, scalar field dynamics, and the theory of cosmological perturbations—both at linear and nonlinear orders. The detailed derivations of perturbation spectra, predictions for the CMB, and the rigorous connections established between soft limits, asymptotic symmetries, and memory effects elucidate both the robust predictions and open questions in inflationary cosmology. The implications for both observational cosmology and theoretical approaches to quantum gravity are profound; as data improves, the constraints on inflationary microphysics and potential high-energy completions will become ever more stringent, delineating the structure of the early universe.