Single-Field Inflation and the Local Ansatz: Distinguishability and Consistency

TL;DR

This paper investigates the relationship between single-field inflation consistency conditions and the local ansatz, showing they cannot be fully satisfied simultaneously and proposing a new observational test for these inflation models.

Contribution

It establishes a fundamental difference between single-field and multifield inflation predictions and introduces a novel method to test the consistency relations observationally.

Findings

Single-field consistency conditions cannot be satisfied by a general local ansatz.

Deviations between the models appear at order $(n_s-1)^2$.

A new scheme for testing inflation consistency relations is proposed.

Abstract

The single-field consistency conditions and the local ansatz have played separate but important roles in characterizing the non-Gaussian signatures of single- and multifield inflation respectively. We explore the precise relationship between these two approaches and their predictions. We demonstrate that the predictions of the single-field consistency conditions can never be satisfied by a general local ansatz with deviations necessarily arising at order $(n_s-1)^2$. This implies that there is, in principle, a minimum difference between single- and (fully local) multifield inflation in observables sensitive to the squeezed limit such as scale-dependent halo bias. We also explore some potential observational implications of the consistency conditions and its relationship to the local ansatz. In particular, we propose a new scheme to test the consistency relations. In analogy with delensing of the cosmic microwave background, one can deproject the coupling of the long wavelength modes with the short wavelength modes and test for residual anomalous coupling.