Cancellation of loop corrections to soft scalar power spectrum

This paper proves that scale-invariant one-loop corrections to superhorizon curvature perturbations from small-scale scalar modes are absent in general inflationary scenarios, including transient ultra-slow-roll, due to dilatation symmetry and explicit one-loop calculations within an in-in effective field theory framework.

The Gist

The Big Picture: The Cosmic Ripple Effect

Imagine the universe during its birth (Inflation) as a giant, stretching rubber sheet. Tiny quantum jitters on this sheet created the seeds for all the galaxies and stars we see today. These jitters are called curvature perturbations.

Usually, physicists believe that once a ripple gets big enough to stretch across the entire observable universe (becoming "superhorizon"), it freezes in place. It becomes a permanent record of the early universe, safe from any new changes happening on tiny, microscopic scales.

The Problem:
Recently, some scientists suggested a scary idea: What if the tiny, chaotic jitters on very small scales could "loop back" and mess up the big, frozen ripples? They claimed that if there was a brief moment of extreme speed (called "Ultra-Slow-Roll") during inflation, these tiny ripples could send a signal back in time to change the big ones. If true, our understanding of the universe's history would be broken because we wouldn't know if the big ripples we see today are the original ones or if they were altered by unknown small-scale physics.

The Solution (This Paper):
The authors of this paper say: "No, that's impossible."

They prove that the big ripples remain perfectly safe. Even if the small-scale physics is wild and crazy, it cannot change the large-scale patterns. The universe has a built-in "cancel button" that ensures the big picture stays clean.

---

The Analogy: The Symphony and the Feedback Loop

To understand how they proved this, let's use an analogy of a Symphony Orchestra.

1. The Big Ripples (The Melody): Imagine the large-scale structure of the universe is the main melody played by the violins. It's slow, steady, and easy to hear.
2. The Small Ripples (The Percussion): The small-scale fluctuations are like the frantic drumming in the back. They are fast, loud, and chaotic.
3. The Fear: Some people worried that the frantic drumming (small scales) was so loud it was creating a "feedback loop" that was distorting the violin melody (large scales), making it sound different than it was originally written.

The Authors' Discovery:
The authors realized that the orchestra has a strict Rule of Conduct (called Dilatation Symmetry). This rule says: "If you change the volume of the whole room, the relationship between the drums and the violins must stay exactly the same."

Because of this rule, the orchestra has a Counter-Tuner (a Counter Term).

• When the drums get too crazy and try to mess up the violins, the Counter-Tuner automatically adds a "negative drum beat" to cancel it out perfectly.
• It's like noise-canceling headphones for the universe. The "noise" from the small scales tries to enter, but the headphones generate an exact opposite wave, and the result is silence (zero change).

The "Magic Trick" of the Proof

The paper uses a mathematical framework called the In-In Formalism (think of it as a special way of calculating how things evolve in time without losing track of cause and effect).

They looked at the "diagrams" (mathematical pictures) of how these ripples interact. They found two types of interactions:

1. The Messy Interaction: The small ripples trying to push the big ones around.
2. The Fix: A specific correction term required by the laws of physics (General Relativity) to keep the universe consistent.

The Punchline:
When you add the "Messy Interaction" and the "Fix" together, they don't just reduce the error; they cancel each other out completely. It's like trying to push a door open while someone else is pulling it shut with exactly the same force. The door doesn't move.

Why Does This Matter?

1. We Can Trust Our Maps: Because the big ripples are immune to the small-scale chaos, the Cosmic Microwave Background (the "baby picture" of the universe) is a reliable record. We don't need to worry that unknown physics in the tiny, unobservable corners of the early universe has rewritten our history.
2. Black Holes are Safe: This paper was partly motivated by the search for Primordial Black Holes (PBHs). Some theories suggested that if small ripples got huge, they could form black holes, but that might mess up the big picture. This paper says: "Go ahead, make those black holes! The big picture of the universe won't even notice."
3. Symmetry is King: The proof relies on a fundamental symmetry of the universe (Dilatation Symmetry). It shows that the laws of physics are so robust that they protect the large-scale structure automatically, without us needing to fine-tune anything.

Summary in One Sentence

The universe has a built-in "noise-canceling" mechanism, enforced by the fundamental laws of gravity, which guarantees that the chaotic jitters of the tiny, early universe cannot distort the smooth, large-scale patterns we observe today.

Technical Summary

1. Problem Statement

In modern inflationary cosmology, the conservation of curvature perturbations ($\zeta$) outside the horizon is a fundamental assumption ensuring the robustness of predictions for the Cosmic Microwave Background (CMB) and Large Scale Structure (LSS). However, recent interest in Primordial Black Hole (PBH) formation has focused on scenarios where small-scale perturbations are significantly enhanced, such as during a transient Ultra-Slow-Roll (USR) phase.

A controversial claim emerged in recent literature suggesting that quantum loop corrections from these enhanced small-scale modes could generate scale-invariant corrections to the large-scale (superhorizon) scalar power spectrum. If true, this would imply that large-scale perturbations are not conserved and are sensitive to unknown small-scale dynamics, potentially violating causality or diffeomorphism invariance. While a previous study by the authors [59] proved the absence of such corrections for tensor modes, the status of scalar modes remained debated. This paper aims to rigorously prove the absence of scale-invariant one-loop corrections to the scalar power spectrum in a general inflationary setup, including transient USR.

2. Methodology

The authors employ a combination of symmetry arguments and explicit perturbative calculations using the in-in formalism (Schwinger-Keldysh formalism) in the Keldysh $r/a$-basis.

• Keldysh $r/a$-basis: The fields are decomposed into "classical" ($r$) and "quantum" ($a$) components ($\phi_r = \frac{1}{2}(\phi_+ + \phi_-)$, $\phi_a = \phi_+ - \phi_-$). This basis simplifies the classification of Feynman diagrams and makes causality manifest (e.g., $a$-$a$ propagators vanish).
• Soft Effective Field Theory (EFT): The authors construct an EFT for soft curvature modes ($\zeta_s$) by integrating out hard modes ($\chi$). They analyze the structure of this EFT on the in-in contour.
• Symmetry Analysis: The core of the proof relies on dilatation symmetry (a remnant of diffeomorphism invariance in the $\zeta$-gauge). The transformation is defined as $\zeta_s(\tau, \mathbf{x}) \to \zeta_s(\tau, \mathbf{x}) + \lambda$ and $\mathbf{x} \to e^{-\lambda}\mathbf{x}$.
• Tadpole Cancellation: The definition of $\zeta$ as a fluctuation around a background requires that its one-point function (tadpole) vanishes. The authors enforce this condition by introducing a counter-term.
• Explicit Calculation: To verify the symmetry arguments, the authors perform explicit one-loop calculations for three increasingly complex models:
  1. A spectator scalar field minimally coupled to $\zeta$.
  2. A toy model where the spectator field is identified with the curvature perturbation itself (without full GR interactions).
  3. The full curvature perturbation with arbitrary interactions, specifically including terms arising from General Relativity (GR) relevant to transient USR.

3. Key Contributions and Theoretical Framework

A. Dilatation Symmetry and Tadpole Cancellation

The authors establish that the soft EFT must respect dilatation symmetry.

• Tadpole Condition: To ensure $\langle \zeta \rangle = 0$, a counter-term $\mathcal{L}_{c.t.} = -\frac{1}{d-1} e^{(d-1)\zeta_s} \Lambda(\tau)$ is required.
• Symmetry Constraint: This counter-term must respect dilatation symmetry. Consequently, the coefficient $\Lambda(\tau)$ is fixed by the tadpole condition, and the counter-term generates specific vertices for higher-order correlations (e.g., two-point functions) with coefficients strictly related to the tadpole coefficient.
• Absence of Fully Retarded Vertices: The combination of tadpole cancellation and dilatation symmetry implies that fully retarded vertices (vertices involving only $a$-type fields in the soft limit, e.g., $\zeta_a \zeta_r^n$) vanish in the soft EFT. This is a crucial structural result.

B. Cancellation Mechanism

The paper demonstrates that scale-invariant corrections arise from specific diagrams involving statistical ($rr$) and retarded ($ra$) propagators.

• The relevant one-loop diagrams involve a loop of hard modes connecting to soft external lines.
• Due to the vanishing of fully retarded vertices, the only surviving contributions come from diagrams where the soft limit of the bulk-to-boundary propagator is time-independent (statistical) or constant (retarded).
• The authors derive a key identity (Eq. 1.1 / Eq. 4.15) relating the derivative of the statistical propagator to a product of statistical and retarded propagators:
  $$\frac{\partial}{\partial \log l} G_{rr}(l) \propto \int d\tau [G_{rr} G_{ra} + G_{ra} G_{rr}]$$
  This identity allows the "shrinking" of loop diagrams into contact terms.
• The Cancellation: The one-loop diagrams generated by the interaction vertices are exactly canceled by the two-point contribution from the dilatation-symmetric counter-term. The coefficient of the counter-term is fixed by the tadpole condition, ensuring the cancellation is automatic and not fine-tuned.

C. Distinction of Scale-Invariant Sources

The authors distinguish between two types of potential scale-invariant corrections:

1. Integrand-level scale-invariance: Terms that are scale-invariant before the conformal time integration. The paper proves these are absent due to the symmetry/cancellation mechanism.
2. Integral-level scale-invariance: Terms that become scale-invariant only after integration (e.g., from the $\tau \to -\infty$ region). The authors argue these are irrelevant for transient USR scenarios because the relevant dynamics occur at finite times, and the $\tau \to -\infty$ region corresponds to standard slow-roll where no enhancement occurs.

4. Results

• General Proof: The paper proves that in any inflationary setup where the initial and final phases are slow-roll, the one-loop correction to the superhorizon scalar power spectrum from small-scale enhanced modes vanishes in the soft limit.
• Explicit Verification:
  • Spectator Field: The cancellation between the loop diagram and the counter-term was explicitly shown for a spectator field.
  • Curvature Perturbation (Toy Model): The cancellation was verified for the curvature perturbation itself, handling the specific kinetic and gradient terms.
  • Full GR Interactions: The most significant result is the extension to full GR interactions (including the cubic terms $\zeta'^3$ and $\zeta (\partial \zeta)^2$ relevant to USR). The authors showed that even with these complex non-linearities, the dilatation-symmetric counter-term ensures the total one-loop correction vanishes.
• Robustness: The result holds regardless of the specific details of the transient USR evolution, provided the symmetry is preserved and the regularization scheme (e.g., dimensional regularization) respects dilatation invariance.

5. Significance

• Resolution of Controversy: This work definitively resolves the debate regarding scale-invariant loop corrections in transient USR scenarios. It refutes claims that large-scale perturbations are sensitive to small-scale PBH-forming dynamics, thereby preserving the standard prediction that superhorizon curvature perturbations are conserved.
• Causality and Symmetry: The result reinforces the deep connection between diffeomorphism invariance (manifested as dilatation symmetry in the $\zeta$-gauge) and causality. The strict cancellation is a direct consequence of the underlying symmetries of General Relativity.
• Methodological Advance: The paper provides a powerful, symmetry-based framework for analyzing loop corrections in inflation. By utilizing the Keldysh $r/a$-basis and identifying the role of the tadpole counter-term, it offers a more robust and general approach than previous methods that relied on solving specific mode equations or invoking Maldacena's consistency relations (though the authors note their method is equivalent to the consistency relation in spirit).
• Implications for PBHs: The findings imply that while small-scale enhancements can produce PBHs and induced gravitational waves, they do not "contaminate" the large-scale CMB power spectrum via quantum loops, ensuring the safety of standard inflationary predictions for large scales.

In summary, Ema et al. demonstrate that the conservation of the curvature perturbation is protected by dilatation symmetry and tadpole cancellation, ensuring that quantum loop corrections from small-scale enhancements do not generate scale-invariant corrections to the large-scale scalar power spectrum.