⚛️ phenomenology

Non-Relativistic Cosmological Collider Signals

This paper proposes a non-relativistic cosmological collider framework where a massive tilted-ghost spectator field generates distinctive squeezed-limit non-Gaussianity through propagation-induced mode deformations rather than boost-breaking interactions, effectively mimicking boostless collider signals via a unique chemical-potential-like mechanism.

Original authors: Matheus C. Ferreira, F. T. Falciano

Published 2026-08-21

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Matheus C. Ferreira, F. T. Falciano

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

In the earliest moments of the universe, a fraction of a second after the Big Bang, space itself expanded at a breathtaking pace. This period, known as inflation, stretched the quantum fluctuations of the vacuum into vast, cosmic structures. While this rapid expansion smoothed out the universe, it also acted as a high-energy laboratory, capable of creating particles that are far too heavy to be produced in any terrestrial particle accelerator. Physicists have long suspected that these heavy particles left behind a faint, unique fingerprint in the distribution of matter we see today. This fingerprint, known as primordial non-Gaussianity, is a subtle statistical pattern that reveals the mass and spin of the particles that existed during inflation. By studying these patterns, scientists hope to perform a kind of "cosmological collider" experiment, using the entire universe as a detector to probe energy scales that are otherwise inaccessible.

The challenge has always been that these signals are incredibly faint. In standard models of inflation, the production of heavy particles is naturally suppressed, making their signatures difficult to distinguish from the background noise of the expanding universe. Researchers have proposed various ways to amplify these signals, often by introducing new forces or breaking specific symmetries in the laws of physics. However, a new study by Matheus C. Ferreira and Felipe T. Falciano suggests a different path. Instead of relying on complex new interactions to boost the signal, they propose that the particles themselves might have behaved in a fundamentally different way as they moved through the early universe. Specifically, they investigated a scenario where a heavy, invisible particle—acting as a spectator to the main inflationary process—did not follow the standard rules of relativity.

The researchers focused on a theoretical particle known as a "tilted ghost." In the language of physics, a "ghost" is a type of field that behaves differently from ordinary matter, often possessing unusual properties in its energy equations. In this specific model, the particle's motion is governed by a non-relativistic dispersion relation, meaning its speed and energy do not scale in the usual way with its momentum. Imagine a particle that, instead of moving like a wave in water, behaves more like a heavy object rolling down a hill where the friction changes depending on how fast it is going. The authors found that this non-relativistic behavior deforms the way the particle propagates through space-time. As the particle moves, its wave-like nature gets stretched and tilted, creating a distinct asymmetry in how it connects to the rest of the universe.

This deformation turns out to be a powerful lever. The study shows that the tilt in the particle's motion acts like a dial that can either amplify or suppress the cosmological signal, depending on the direction of the tilt. In previous models, the signal from heavy particles was often so weak it was effectively invisible. Here, the researchers demonstrated that the non-relativistic corrections inherent to this tilted-ghost scenario can significantly enhance the signal, making it much easier to detect. The mechanism works because the particle's unusual propagation changes the mathematical coefficients that determine the strength of its interaction with the inflationary background. This is not a result of adding a new force or a new particle, but rather a consequence of how an existing type of particle moves when the rules of relativity are slightly bent.

The team calculated the expected signal in the "squeezed limit," a specific configuration where one part of the cosmic pattern is much larger than the others. In this regime, the signal typically appears as a rhythmic oscillation, a kind of cosmic clock ticking in the data. The study found that while the frequency of this clock remains the same, the amplitude—the loudness of the signal—changes dramatically based on the tilt. If the tilt is in one direction, the signal becomes much stronger, potentially bringing heavy particles within the reach of future observations. If the tilt is in the opposite direction, the signal is suppressed. This asymmetry provides a clear, testable signature that distinguishes this model from other theories.

Crucially, the authors emphasize that this effect arises from the propagation of the particle itself, rather than from new interaction vertices or external chemical potentials that some other theories require. The particle does not need to be charged or coupled to a rolling background field to produce this effect; the non-relativistic nature of its movement is sufficient. The study also clarifies that this enhancement is not unlimited. It works best for particles within a certain mass range and requires the tilt to be within a controlled, perturbative regime. If the tilt becomes too large, the theoretical framework breaks down, and the particle would become unstable. However, within these safe boundaries, the model offers a simple and natural explanation for how the universe could have amplified the whispers of heavy particles.

By isolating this mechanism, the researchers provide a new tool for interpreting data from the cosmic microwave background and large-scale structure surveys. If future observations detect a cosmological collider signal with an amplitude that does not match standard predictions, it could point directly to this kind of non-relativistic propagation. The work suggests that the universe might be hiding its most massive secrets not in the complexity of its interactions, but in the subtle, non-relativistic ways its fundamental fields move. This shifts the focus of the search, encouraging scientists to look for specific patterns of enhancement and suppression that arise purely from the kinematics of the early universe. The findings offer a fresh perspective on how the microscopic physics of the Big Bang might be encoded in the macroscopic structure of the cosmos today.

Technical Summary: Non-relativistic cosmological collider signals

Problem Statement
The paper addresses the generation of primordial non-Gaussianity (PNG) in the context of the "Cosmological Collider" program. Standard cosmological collider signals arise from the exchange of massive fields during inflation, imprinting characteristic non-analytic, oscillatory signatures in the squeezed limit of the bispectrum. However, these signals are typically exponentially suppressed by a Boltzmann factor $e^{-\pi \mu}$ (where $\mu \sim m/H$ ) for heavy particles. While previous work has proposed mechanisms to alleviate this suppression—such as non-Bunch-Davies initial states, chemical potentials induced by rolling backgrounds, or "boostless" scenarios involving sound-speed hierarchies—this work investigates a distinct, non-relativistic realization. Specifically, the authors explore how a massive spectator field with a "tilted-ghost" dispersion relation modifies the collider signal. Unlike standard boostless constructions where boost-breaking occurs primarily at interaction vertices, this framework posits that the signal is modified fundamentally by the propagation of the spectator mode itself.

Methodology
The authors construct a minimal effective field theory (EFT) model consisting of a standard slow-roll inflaton $\phi$ and a massive scalar spectator $\sigma$ with a tilted-ghost dispersion relation. The background is approximated as quasi-de Sitter space.

Model Construction: The spectator field is governed by a modified dispersion relation incorporating a quartic momentum term ( $\gamma^2 k^4 \eta^2$ ) and a quadratic relativistic correction ( $\pm \beta k^2$ ). The parameter $\gamma = H/M$ sets the hierarchy between the Hubble scale and the ghost scale $M$ , while the signed parameter $\lambda = s_{\text{tilt}}\beta/(4\gamma)$ controls the tilt.
Mode Functions: The authors derive the exact analytic solution for the spectator mode functions. By transforming the equation of motion, they identify the solution as a Whittaker function $W_{\kappa, \mu}(z)$ with a specific argument $z = i\gamma k^2 \eta^2$ . The early-time (UV) vacuum is fixed by the pure ghost regime ( $k^4$ dominance), selecting a specific WKB branch.
Dynamics and Crossover: The paper analyzes the semiclassical (WKB) regimes, identifying a crossover scale where the relativistic $k^2$ term becomes relevant relative to the quartic term. For light fields, the positive-tilt branch can exhibit a transient tachyonic instability before freeze-out. For heavy (principal series) fields, the tilt does not induce a real turning point in the controlled regime but instead acts as a branch-asymmetry parameter, biasing the relative weights of the late-time oscillatory branches.
Cosmological Collider Calculation: Using the Schwinger-Keldysh (in-in) formalism, the authors compute the tree-level exchange bispectrum generated by the interaction of the inflaton with the spectator. They isolate the non-analytic "soft" contribution (where the exchanged momentum $k_3 \to 0$ ) from the analytic "hard" background contributions.

Key Contributions and Results

Non-Relativistic Propagation Mechanism: The primary result is that the cosmological collider signal is modified not by new interaction vertices, but by the deformation of the spectator's mode functions due to non-relativistic propagation. The tilted-ghost dispersion relation induces a "tilt" in the mode functions that acts as an effective branch-asymmetry parameter.
Modified Squeezed-Limit Signal: The resulting squeezed-limit bispectrum retains the characteristic logarithmic oscillatory structure $\cos[\mu \log(k_3/k_1) + \delta]$ $cos [μ lo g (k_{3} / k_{1}) + δ]$ , but with distinct modifications:
- Amplitude: The amplitude is dressed by Whittaker connection coefficients that depend on the signed tilt $\lambda$ . Depending on the sign of $\lambda$ , the signal can be significantly enhanced or suppressed relative to the pure-ghost or standard de Sitter cases.
- Phase: The tilt induces a phase shift in the oscillatory signal, effectively displacing the clock signal in log-momentum space.
- Normalization: The signal is normalized to a pure-ghost WKB vacuum rather than the standard relativistic Bunch-Davies vacuum, leading to different ultraviolet normalizations compared to standard Whittaker templates.
Comparison with Standard Templates: The authors demonstrate that the tilted-ghost signal cannot be obtained by a simple reparameterization of standard boostless or chemical-potential templates.
- Unlike chemical-potential models, there is no conserved charge or external energy scale sourcing the heavy particles; the effect arises purely from the kinematic projection of the UV ghost branch onto the IR oscillatory branches.
- Unlike standard Whittaker modes (which reduce to Hankel functions in the $\lambda \to 0$ limit), the tilted-ghost solution reduces to a pure-ghost WKB mode.
- The sign of the tilt $\lambda$ produces an asymmetric response: the $\lambda < 0$ branch is enhanced, while $\lambda > 0$ is suppressed, a behavior distinct from standard chemical-potential deformations.
Enhancement vs. Suppression: The paper shows that for moderate masses, the tilted-ghost mechanism can enhance the signal amplitude by orders of magnitude compared to standard de Sitter exchange (e.g., $f_{\text{NL}}^{\text{osc}} \sim \mathcal{O}(1)$ for $\mu=1$ ), though it does not generically achieve the parametric reach of chemical-potential models for very heavy masses ( $m \gg H$ ).

Significance and Claims
The paper claims to provide a simple effective framework where boostless-collider phenomenology and branch asymmetries arise naturally from non-relativistic propagation effects. The significance lies in:

Distinct Dynamical Origin: It establishes that collider signals can be generated and modified by the propagation of the exchanged field itself, rather than solely by boost-breaking interactions.
New Observables: The model predicts a specific signature where the amplitude and phase of the clock signal are controlled by a signed tilt parameter $\lambda$ , offering a potential observational discriminant against standard relativistic or chemical-potential scenarios.
Effective Description: The authors position the work as an effective description of tilted-ghost inflation, valid when the spectator energy density is subdominant. They explicitly note that a full UV completion and detailed analysis of backreaction for large tilts are beyond the scope of the current work.

The authors conclude that while the tilted-ghost scenario improves the visibility of heavy particles relative to ordinary de Sitter exchange, it operates through a mechanism distinct from chemical potentials or non-Bunch-Davies initial states, relying instead on the non-relativistic projection of the UV vacuum onto the late-time oscillatory branches.

Technical Summary: Non-relativistic cosmological collider signals

More like this