General Scalar Field Inflation ACT Attractors: Utilizing the $n_s(r)$ relation
V. K. Oikonomou
TL;DR
This work addresses the tension between ACT constraints and traditional inflationary models by adopting a top-down, model-agnostic approach: specify the scalar spectral index as a function of the tensor-to-scalar ratio, $n_s=f(r)$, and reconstruct the underlying potential $V(φ)$ by solving the resulting differential equation. Building on plateau-like attractors that yield a universal relation $n_s(r) \approx 1 - α r^{1/2}$, the authors analyze several simple yet tractable forms of $f(r)$, deriving explicit analytic potentials and slow-roll predictions. Four models (Models I–IV) are developed, with three main classes of $n_s(r)$ attractors based on $n_s(r)=γ ± β r ± \sqrt{r}$, each yielding closed-form $V(φ)$ and confronting ACT and Planck/BICEP constraints; viable parameter ranges are identified for ACT compatibility. The results demonstrate a viable pathway to ACT-compatible, plateau-like inflation within a top-down framework and point to further exploration, including potential embedding in no-scale supergravity theories.
Abstract
The ACT data have severely constrained the single scalar field models. Known models of inflation, like the Starobinsky model, the Higgs model and the $a$-attractors are at least $2σ$ off the ACT data. In this work we aim to provide a top-to-bottom approach in single scalar field inflationary cosmology compatible with the ACT data. Specifically, inspired by the fact that the Starobinsky model, the Higgs model and the $a$-attractors, all being plateau potentials, result to the same attractor relation between the spectral index of scalar perturbations and the tensor-to-scalar ratio, which is of the form $n_s(r)=1-αr^{1/2}$, in this work we seek for attractors of the form $n_s(r)=f(r)$ that may lead to ACT-compatible inflation. Specifically, we fix the function $f(r)$ to have a specific desirable form and then solve the differential equation $n_s(r)=f(r)$ to find the potential which results to the relation $n_s(r)=f(r)$. We discovered analytically three classes of potentials which are variants of the general form $n_s(r)=γ\pm βr \pm r^{1/2}$ and all these models are found to be compatible with the ACT data.
