Stochastic Constant-Roll Inflation

• Stochastic constant-roll inflation is a framework that models inflationary dynamics with a constant second slow-roll parameter using Langevin and Fokker–Planck methods.
• The spectral analysis of the Fokker–Planck operator enables an exact treatment of first-passage time statistics and reveals the interplay between classical drift and stochastic diffusion.
• Rare hilltop-crossing trajectories significantly alter the inflation duration, impacting the curvature perturbation tail and the conditions for primordial black-hole formation.

Searching arXiv for papers on stochastic constant-roll inflation and closely related stochastic-inflation formalisms. Stochastic constant-roll inflation is the stochastic treatment of inflationary dynamics in backgrounds for which the second slow-roll parameter is constant, $$\epsilon_2 \equiv d\ln \epsilon_1/dN = \mathrm{const}$$. In this setting, the long-wavelength inflaton evolves under classical drift and quantum diffusion, and the resulting perturbation statistics can be analyzed with Langevin and Fokker–Planck methods, or, in pure constant roll, with a closed-form nonlinear $\Delta N$ map. For a quadratic hilltop potential,

$V(\phi)=V_0-\tfrac12 m^2\phi^2,$

the growing attractor obeys

$$\phi_{\rm cl}(N)=\phi_0 e^{A_+N},\qquad A_+=\tfrac32\Bigl(\sqrt{1-\tfrac43\eta_V}-1\Bigr),$$

so that $\epsilon_2=2A_+$ is constant. Recent work has shown that, once one allows trajectories to cross the hilltop and encounter a reflecting boundary on the far side, rare but very long-lived histories qualitatively alter first-passage statistics and the tail structure of the coarse-grained $\Delta N$ distribution; earlier work had already derived an exact non-Gaussian curvature-perturbation distribution in pure constant roll and connected it to primordial black-hole production, while Pattison et al. established the validity of the stochastic formalism beyond slow roll and quantified gauge corrections to the noise amplitude (Tomberg, 8 Jun 2026, Tomberg, 2023, Pattison et al., 2019).

1. Foundational stochastic framework beyond slow roll

Pattison et al. formulated the stochastic dynamics of a single inflaton field beyond the slow-roll approximation in the uniform-$N$ gauge, where the number of e-folds $N\equiv \ln a$ is unperturbed. The exact phase-space Langevin system is

$$\frac{d\phi}{dN}=\pi+\xi_\phi(N),\qquad \frac{d\pi}{dN}=-[3-\epsilon_1(\phi,\pi)]\,\pi-\frac{V_{,\phi}(\phi)}{H^2(\phi,\pi)}+\xi_\pi(N),$$

with $\epsilon_1\equiv \pi^2/(2M_{\rm p}^2)$ and Gaussian white noises satisfying

$\Delta N$0

In many practical cases one reduces this system to a one-field Langevin equation,

$\Delta N$1

where $\Delta N$2 is the classical drift and $\Delta N$3 is the field-only diffusion coefficient (Pattison et al., 2019).

A key technical issue is that the noise amplitude is usually computed from the mode function of the Mukhanov–Sasaki variable in the spatially-flat gauge, whereas the Langevin equation is written in uniform-$\Delta N$4 gauge. Pattison et al. showed that the corresponding gauge correction can be written as

$\Delta N$5

and found that for attractor backgrounds the correction satisfies $\Delta N$6. They also demonstrated that the separate-universe approach remains valid on super-Hubble scales without any slow-roll assumption: each patch obeys the homogeneous Klein–Gordon equation up to $\Delta N$7 corrections. This establishes the formal basis for applying stochastic inflationary methods to constant-roll regimes rather than only to slow-roll backgrounds (Pattison et al., 2019).

2. Constant-roll realization near a quadratic hilltop

In the hilltop model emphasized by Tomberg, the local potential maximum at $\Delta N$8 is characterized by

$\Delta N$9

In an almost constant-Hubble regime, $V(\phi)=V_0-\tfrac12 m^2\phi^2,$0 and $V(\phi)=V_0-\tfrac12 m^2\phi^2,$1, the homogeneous inflaton equation becomes

$V(\phi)=V_0-\tfrac12 m^2\phi^2,$2

The growing attractor solution,

$V(\phi)=V_0-\tfrac12 m^2\phi^2,$3

implies

$V(\phi)=V_0-\tfrac12 m^2\phi^2,$4

which is precisely the constant-roll condition. Along this attractor, the stochastic long mode obeys

$V(\phi)=V_0-\tfrac12 m^2\phi^2,$5

with linear drift

$V(\phi)=V_0-\tfrac12 m^2\phi^2,$6

and constant diffusion amplitude $V(\phi)=V_0-\tfrac12 m^2\phi^2,$7 times a constant-roll correction factor (Tomberg, 8 Jun 2026).

This linear-drift, constant-diffusion specialization is the technically decisive simplification behind the later spectral analysis. It permits an exact characterization of the relevant Fokker–Planck operator in a bounded domain with one absorbing and one reflecting boundary, and it sharply separates the classical outward drift from the diffusion-driven possibility of hilltop crossing. The latter possibility is the source of the qualitative differences between beyond-hilltop stochastic solutions and treatments that restrict trajectories to a single side of the hilltop (Tomberg, 8 Jun 2026).

3. Fokker–Planck operator and spectral solution

For the one-field constant-roll system, the probability density $V(\phi)=V_0-\tfrac12 m^2\phi^2,$8 satisfies

$V(\phi)=V_0-\tfrac12 m^2\phi^2,$9

Tomberg imposed an absorbing boundary at $$\phi_{\rm cl}(N)=\phi_0 e^{A_+N},\qquad A_+=\tfrac32\Bigl(\sqrt{1-\tfrac43\eta_V}-1\Bigr),$$0, interpreted as the end-of-inflation field value, and a reflecting boundary at $$\phi_{\rm cl}(N)=\phi_0 e^{A_+N},\qquad A_+=\tfrac32\Bigl(\sqrt{1-\tfrac43\eta_V}-1\Bigr),$$1, mimicking a steep classical slope or second minimum. The reflecting condition is a vanishing probability current,

$$\phi_{\rm cl}(N)=\phi_0 e^{A_+N},\qquad A_+=\tfrac32\Bigl(\sqrt{1-\tfrac43\eta_V}-1\Bigr),$$2

while the absorbing condition is

$$\phi_{\rm cl}(N)=\phi_0 e^{A_+N},\qquad A_+=\tfrac32\Bigl(\sqrt{1-\tfrac43\eta_V}-1\Bigr),$$3

The solution is sought in separated form,

$$\phi_{\rm cl}(N)=\phi_0 e^{A_+N},\qquad A_+=\tfrac32\Bigl(\sqrt{1-\tfrac43\eta_V}-1\Bigr),$$4

where the eigenfunctions satisfy

$$\phi_{\rm cl}(N)=\phi_0 e^{A_+N},\qquad A_+=\tfrac32\Bigl(\sqrt{1-\tfrac43\eta_V}-1\Bigr),$$5

Introducing the weight function

$$\phi_{\rm cl}(N)=\phi_0 e^{A_+N},\qquad A_+=\tfrac32\Bigl(\sqrt{1-\tfrac43\eta_V}-1\Bigr),$$6

and setting $$\phi_{\rm cl}(N)=\phi_0 e^{A_+N},\qquad A_+=\tfrac32\Bigl(\sqrt{1-\tfrac43\eta_V}-1\Bigr),$$7, the operator can be rewritten in self-adjoint form $$\phi_{\rm cl}(N)=\phi_0 e^{A_+N},\qquad A_+=\tfrac32\Bigl(\sqrt{1-\tfrac43\eta_V}-1\Bigr),$$8. Sturm–Liouville theory then guarantees a discrete spectrum

$$\phi_{\rm cl}(N)=\phi_0 e^{A_+N},\qquad A_+=\tfrac32\Bigl(\sqrt{1-\tfrac43\eta_V}-1\Bigr),$$9

with orthonormal eigenfunctions $\epsilon_2=2A_+$0 and completeness relation

$\epsilon_2=2A_+$1

In practice, Tomberg solved the eigenvalue problem by a shooting method, requiring exactly $\epsilon_2=2A_+$2 nodes in $\epsilon_2=2A_+$3 while enforcing the two boundary conditions (Tomberg, 8 Jun 2026).

The methodological significance is that the spectral method yields the full late-time probability density and, through the same eigensystem, the full first-passage-time statistics. This contrasts with approximations that focus only on low moments or assume that the large-$\epsilon_2=2A_+$4 tail remains purely exponential all the way to arbitrarily large $\epsilon_2=2A_+$5 (Tomberg, 8 Jun 2026).

4. First-passage statistics and the failure of the mean background

The first-passage-time distribution through the absorbing boundary is given by the boundary current,

$\epsilon_2=2A_+$6

with spectral decomposition

$\epsilon_2=2A_+$7

where

$\epsilon_2=2A_+$8

The mean first-passage time is

$\epsilon_2=2A_+$9

In the wide-boundary limit, $\Delta N$0 and $\Delta N$1 in suitable units, the smallest eigenvalue satisfies $\Delta N$2, so the mean diverges as $\Delta N$3. Tomberg showed that this divergence is dominated by extremely rare trajectories that cross the hilltop, get stuck near the reflecting boundary on the other side, and then tunnel out slowly in a way dictated by the lowest eigensolution. Although rare, these trajectories dominate the global average, so the mean does not properly describe the inflationary background (Tomberg, 8 Jun 2026).

Tomberg therefore proposed using the median rather than the mean. The median $\Delta N$4 satisfies

$\Delta N$5

or equivalently

$\Delta N$6

At moderate barrier heights this relation is dominated by the first few eigenterms, and for initial $\Delta N$7 on the inflationary slope, $\Delta N$8, the median tracks the classical value

$\Delta N$9

almost exactly. A common simplification is to identify the background e-fold number with the mean; in the beyond-hilltop stochastic problem, Tomberg argued that the median is the more robust background measure for the $N$0 formalism because it reflects the bulk of realizations rather than rare outliers (Tomberg, 8 Jun 2026).

5. Structure of the coarse-grained $N$1 distribution

In the coarse-grained $N$2 formalism, one selects a coarse-graining time $N$3, reads off the stochastic field value $N$4, and adds the remaining first-passage time $N$5 to the absorbing boundary. Defining a single background by the median prescription,

$N$6

one obtains a one-to-one map $N$7. For $N$8, the average remaining time is approximately classical,

$N$9

and the probability density $N\equiv \ln a$0 is nearly Gaussian of width $N\equiv \ln a$1. The associated small-$N\equiv \ln a$2 distribution is the modified Gaussian

$N\equiv \ln a$3

This gives a nearly Gaussian core and an exponential tail (Tomberg, 8 Jun 2026).

For $N\equiv \ln a$4, the behavior changes qualitatively. The median remaining time saturates at a constant,

$N\equiv \ln a$5

which is the largest allowed $N\equiv \ln a$6. The distribution then piles up into a skewed plateau and finally a sharp peak at $N\equiv \ln a$7. Quantitatively, Tomberg found a plateau height of order

$N\equiv \ln a$8

and ultimately a $N\equiv \ln a$9-function-like spike at $$\frac{d\phi}{dN}=\pi+\xi_\phi(N),\qquad \frac{d\pi}{dN}=-[3-\epsilon_1(\phi,\pi)]\,\pi-\frac{V_{,\phi}(\phi)}{H^2(\phi,\pi)}+\xi_\pi(N),$$0. On a log plot, the full $$\frac{d\phi}{dN}=\pi+\xi_\phi(N),\qquad \frac{d\pi}{dN}=-[3-\epsilon_1(\phi,\pi)]\,\pi-\frac{V_{,\phi}(\phi)}{H^2(\phi,\pi)}+\xi_\pi(N),$$1 displays a Gaussian peak around small $$\frac{d\phi}{dN}=\pi+\xi_\phi(N),\qquad \frac{d\pi}{dN}=-[3-\epsilon_1(\phi,\pi)]\,\pi-\frac{V_{,\phi}(\phi)}{H^2(\phi,\pi)}+\xi_\pi(N),$$2, an intermediate approximate exponential tail $$\frac{d\phi}{dN}=\pi+\xi_\phi(N),\qquad \frac{d\pi}{dN}=-[3-\epsilon_1(\phi,\pi)]\,\pi-\frac{V_{,\phi}(\phi)}{H^2(\phi,\pi)}+\xi_\pi(N),$$3 and then $$\frac{d\phi}{dN}=\pi+\xi_\phi(N),\qquad \frac{d\pi}{dN}=-[3-\epsilon_1(\phi,\pi)]\,\pi-\frac{V_{,\phi}(\phi)}{H^2(\phi,\pi)}+\xi_\pi(N),$$4, a flattening to a plateau once the first tunneling mode overtakes the classical modes, and a final spike near $$\frac{d\phi}{dN}=\pi+\xi_\phi(N),\qquad \frac{d\pi}{dN}=-[3-\epsilon_1(\phi,\pi)]\,\pi-\frac{V_{,\phi}(\phi)}{H^2(\phi,\pi)}+\xi_\pi(N),$$5 (Tomberg, 8 Jun 2026).

These results directly modify the standard expectation that stochastic inflation near a hilltop produces only a Gaussian core plus a single exponential tail. Tomberg argued that similar intricacies should arise in primordial-black-hole models with a shallow secondary minimum, because patches that cross into a diffusion-dominated well can inflate for almost exactly a maximal duration before escaping (Tomberg, 8 Jun 2026).

6. Curvature perturbations, deterministic $$\frac{d\phi}{dN}=\pi+\xi_\phi(N),\qquad \frac{d\pi}{dN}=-[3-\epsilon_1(\phi,\pi)]\,\pi-\frac{V_{,\phi}(\phi)}{H^2(\phi,\pi)}+\xi_\pi(N),$$6, and primordial black holes

Tomberg’s earlier treatment of stochastic constant-roll inflation and primordial black holes solved the system along the classical constant-roll trajectory by drawing stochastic kicks from a numerically computed power spectrum, beyond the usual de Sitter approximation. The integrated curvature power spectrum is

$$\frac{d\phi}{dN}=\pi+\xi_\phi(N),\qquad \frac{d\pi}{dN}=-[3-\epsilon_1(\phi,\pi)]\,\pi-\frac{V_{,\phi}(\phi)}{H^2(\phi,\pi)}+\xi_\pi(N),$$7

and in pure constant roll, with $$\frac{d\phi}{dN}=\pi+\xi_\phi(N),\qquad \frac{d\pi}{dN}=-[3-\epsilon_1(\phi,\pi)]\,\pi-\frac{V_{,\phi}(\phi)}{H^2(\phi,\pi)}+\xi_\pi(N),$$8 and $$\frac{d\phi}{dN}=\pi+\xi_\phi(N),\qquad \frac{d\pi}{dN}=-[3-\epsilon_1(\phi,\pi)]\,\pi-\frac{V_{,\phi}(\phi)}{H^2(\phi,\pi)}+\xi_\pi(N),$$9,

$\epsilon_1\equiv \pi^2/(2M_{\rm p}^2)$0

Writing the stochastic solution as

$\epsilon_1\equiv \pi^2/(2M_{\rm p}^2)$1

one finds that $\epsilon_1\equiv \pi^2/(2M_{\rm p}^2)$2 is Gaussian with

$\epsilon_1\equiv \pi^2/(2M_{\rm p}^2)$3

The nonlinear $\epsilon_1\equiv \pi^2/(2M_{\rm p}^2)$4 formula gives the coarse-grained curvature perturbation

$\epsilon_1\equiv \pi^2/(2M_{\rm p}^2)$5

which implies

$\epsilon_1\equiv \pi^2/(2M_{\rm p}^2)$6

and therefore

$\epsilon_1\equiv \pi^2/(2M_{\rm p}^2)$7

For $\epsilon_1\equiv \pi^2/(2M_{\rm p}^2)$8 this is approximately Gaussian with variance $\epsilon_1\equiv \pi^2/(2M_{\rm p}^2)$9, whereas for $\Delta N$00 it develops an exponential tail,

$\Delta N$01

In this regime, the perturbation distribution is controlled by the two parameters $\Delta N$02 (Tomberg, 2023).

The same work connected the non-Gaussian tail to primordial-black-hole formation. With collapse threshold $\Delta N$03, the PBH mass at scale $\Delta N$04 is

$\Delta N$05

and the collapsing fraction is

$\Delta N$06

Numerically, for asteroid-mass PBHs with $\Delta N$07, $\Delta N$08, and $\Delta N$09, pure slow roll requires $\Delta N$10, whereas for $\Delta N$11 one can reduce this to

$\Delta N$12

an $\Delta N$13-fold reduction. Tomberg also showed that a non-stochastic $\Delta N$14 treatment with a single initial Gaussian field kick reproduces precisely the same $\Delta N$15 in the pure constant-roll regime, but that the stochastic framework clarifies that the kicks actually arrive continuously over the constant-roll phase and that outside pure constant roll the deterministic shortcut no longer applies (Tomberg, 2023).