UC San Diego

Cosmic inflation is the leading theory for the generation of fluctuations in the early universe. Proposed interactions during inflation can lead to the generation of novel signatures on top of the minimal Gaussian, nearly scale-invariant fluctuations predicted by the standard model of cosmology. While the fields generating these signatures have long since decayed and are no longer relevant for the dynamics of the late-time universe, their interactions with the inflaton leave perturbations in the spacetime curvature that seed the matter distribution of the present-day universe through gravitational interactions, allowing the large-scale structure (LSS) to be a laboratory for probing these primordial interactions at the present day. In this dissertation, we study the non-Gaussian signals generated by field interactions during inflation by analyzing the scale-dependent features they imprint in the power spectrum of galaxies, and the degeneracies with nonlinear evolution and other late-time effects. In Chapter 1, we discuss the theoretical background and motivation of this work. In Chapter 2, we study the scale-dependent bias that results from interactions involving light fields during inflation, and explore the feasibility of detecting such signatures from present and future galaxy surveys. We show that the survey sensitivity can be significantly improved by optimizing the survey design. In Chapter 3, we expand the search further by considering a wider range of scale-dependent signatures that are generated through the cosmological collider. We make measurements and constraints on the non-Gaussian amplitude in the expanded parameter space. While no signal has been detected, our analysis offers insights into future galaxy surveys’ ability to constrain inflationary models through the two-point statistics. We further discuss the effects of incorporating additional physical effects in our modeling on the constraints.