Constraining Inflation with Little Red Dots

Dionysios Karagiannis, P. Daniel Meerburg

First listed 2026-09-23 | Last updated 2026-09-22

Abstract

Primordial non-Gaussianity is a direct probe of the physics of inflation. The tightest bounds still come from the cosmic microwave background, but large-scale structure is rapidly catching up: its intrinsically three-dimensional maps contain far more modes, provided the tracer combines large volume, high redshift, strong bias and sufficient number density. The Little Red Dots (LRDs) discovered by JWST plausibly offer all four. Inspired by recent work on BAO measurements using LRDs we forecast the joint galaxy power spectrum and bispectrum of a $14{,}000\,{\rm deg}^2$ LRD spectroscopic survey over $4<z<9$, marginalising over $32$ nuisance parameters and including the non-Gaussian bispectrum covariance. We find $σ(f_{\mathrm{NL}}^{\mathrm{loc}})=0.32$, $σ(f_{\mathrm{NL}}^{\mathrm{equil}})=32$ and $σ(f_{\mathrm{NL}}^{\mathrm{orth}})=10$, improving on Planck by factors of $16$, $1.5$ and $2.3$. The local bound lies a factor of three below the $σ(f_{\mathrm{NL}}^{\mathrm{loc}})\simeq1$ threshold that separates broad classes of single- and multi-field inflation. LRD clustering therefore has the potential to become an important probe of inflation, and makes a strong case for future wide-field near- to mid-infrared spectroscopy.

Short digest

Karagiannis and Meerburg forecast how a 14,000 deg² spectroscopic survey of JWST little red dots across 4<z<9 could constrain primordial non-Gaussianity through the joint redshift-space galaxy power spectrum and bispectrum. Using an HOD-calibrated LRD bias hierarchy, 32 marginalized nuisance parameters, and a non-Gaussian bispectrum covariance, they obtain σ(fNLloc)=0.32, σ(fNLequil)=32, and σ(fNLorth)=10. The local result would improve on Planck by a factor of 16 and fall roughly threefold below the σ(fNLloc)≈1 benchmark separating broad single- and multi-field inflation classes, making LRD clustering a compelling science case for wide near- to mid-infrared spectroscopy. A central forecasting lesson is that non-Gaussian covariance degrades the local constraint by a factor of four to six relative to a Gaussian-diagonal treatment, because squeezed configurations are strongly coupled.

Key figures to inspect

• Figure 1. This is the headline comparison: it places the conservative LRD forecast for local, equilateral, and orthogonal PNG alongside selected CMB and large-scale-structure forecasts, directly showing the claimed improvement over Planck and the crossing of the σ(fNLloc)=1 benchmark.
• Figure 2. This figure identifies which assumptions control the forecast robustness. In particular, it shows that LRD abundance dominates the astrophysical uncertainty budget, while equilateral and orthogonal constraints are especially exposed because their information is carried by shot-noise-limited small-scale triangles.
• Figure 5. This survey-area scaling figure translates the inflation forecast into an observational design requirement. It contrasts Gaussian and non-Gaussian covariance treatments across sky fraction and shows how the local, equilateral, and orthogonal constraints respond differently to widening the spectroscopic footprint.