On Cosmological Correlators with Boundary Contributions

This paper utilizes the cosmological bootstrap framework to establish criteria distinguishing when boundary terms in quasi-de Sitter spacetime yield non-vanishing contributions to cosmological correlators, applying these insights to systematically classify and extract boundary effects in both dS-invariant and boost-breaking massive-exchange scenarios.

The Gist

Imagine the early universe as a giant, expanding balloon. Physicists try to understand what happened inside this balloon (the "bulk") by looking at the patterns left on its surface (the "boundary") after it stopped inflating. These patterns are called cosmological correlators—essentially, snapshots of how different particles were connected to each other during that explosive growth.

For a long time, scientists believed that to understand these patterns, they only needed to study the interactions happening deep inside the balloon. They thought the "edges" of the balloon (the boundary) were just empty space where nothing interesting happened, or that any effects from the edge were just mathematical tricks that could be ignored.

The Big Idea of This Paper
The authors of this paper say: "Wait a minute. The edge matters."

They argue that the boundary isn't just a passive wall; it actively contributes to the patterns we see. Sometimes, what happens at the very end of inflation leaves a permanent mark that cannot be erased by looking only at the middle of the universe.

Here is how they explain it using simple analogies:

1. The "Redundant Moves" Analogy (Field Redefinitions)

In physics, you can often describe the same situation in different ways. Imagine you are playing a game of chess. You could describe a move as "moving the pawn forward," or you could describe it as "moving the pawn forward and then immediately renaming the square it landed on." The game state is the same, but the description changed.

In the universe, physicists use "field redefinitions" to simplify their math. They try to rename or reshape the fields (the particles) to make the equations cleaner. Usually, they assume that if a term in the equation looks like it belongs to the "edge" (a boundary term), it's just a result of this renaming and can be thrown away.

The Paper's Discovery:
The authors show that in the expanding universe, this isn't always true. When you "rename" the fields, you don't just change the description; you accidentally leave a physical "stain" on the edge of the universe. It's like if, every time you renamed a chess square, you accidentally left a tiny drop of ink on the board's edge. That ink is real, and it changes the final picture.

2. The "Scalpel" Analogy (Cutting the Diagrams)

To prove this, the authors developed a new set of rules, which they call "diagrammatic reduction rules."

Imagine the interactions between particles as a complex web of strings (Feynman diagrams).

• The Old Way: Scientists would try to untangle the whole web to see the final shape.
• The New Way: The authors use a "scalpel" (mathematical tools called Integration by Parts and Equations of Motion) to cut specific strings in the web.

When they cut a string, two things happen:

1. The Bulk Part: The main part of the web changes, but it's still there.
2. The Boundary Part: The cut string leaves a loose end that snaps onto the edge of the universe.

The paper provides a checklist (Criteria 1, 2, and 3) to tell you when that loose end on the edge is important:

• Criterion 1: Did the cut actually touch the edge? (If the string was cut in the middle of nowhere, it doesn't matter).
• Criterion 2: Is the thing left on the edge heavy or light? (If it's a heavy particle, it might fade away quickly. If it's light, it stays).
• Criterion 3: Is it spinning or moving sideways? (If the leftover piece involves complex sideways motion, it might cancel itself out).

3. The "Heavy vs. Light" Particle Analogy

The paper looks at two types of particles:

• Heavy Particles (The Principal Series): These are like heavy rocks. When they interact, they leave a distinct, sharp mark on the boundary. The authors show that for these, the "edge marks" are real and necessary to get the right answer.
• Light Particles (The Complementary Series): These are like feathers. They are tricky. Sometimes, the "edge marks" from feathers don't cancel out, leading to weird, infinite numbers (divergences) in the math. The authors show how to handle these feathers so the math makes sense.

4. The "Recipe Book" (Recursion)

Finally, the authors realized that instead of cooking every single dish (calculating every possible particle interaction) from scratch, they could use a "recipe book."

They found a pattern: If you know the result for a simple interaction, you can use a specific rule (a recursion relation) to figure out the result for a more complex interaction with more derivatives (more twists and turns in the math). It's like knowing how to bake a basic cake allows you to instantly know how to bake a cake with extra layers, without having to start over.

Summary

In short, this paper tells us that the edge of the inflationary universe is not a silent observer.

• Old view: The edge is just a mathematical artifact; ignore it.
• New view: The edge is a real physical participant. When we simplify our equations, we must account for the "stains" left on the edge.
• The Tool: The authors gave us a new set of "scissors" and a "checklist" to figure out exactly which edge effects are real and which are just noise.

This helps physicists build a more accurate "bootstrap" (a way to build the theory of the universe from the ground up) by ensuring they don't accidentally throw away the most interesting parts of the cosmic story.

Technical Summary

Technical Summary: On Cosmological Correlators with Boundary Contributions

Problem Statement
Cosmological correlators, which probe the early universe and high-energy physics, receive contributions from both bulk field interactions in quasi-de Sitter (dS) spacetime and boundary terms at the end of inflation. While the cosmological bootstrap program has successfully classified bulk interactions using symmetries, locality, and unitarity, the role of boundary terms has often been dismissed. In flat-space scattering amplitudes, terms proportional to equations of motion (EoM) or total derivatives (Integration by Parts, IBP) are redundant; they either vanish due to vacuum boundary conditions at infinity or can be removed via field redefinitions without affecting the S-matrix. However, in dS spacetime, the presence of a future spacelike boundary (the reheating surface) invalidates the standard flat-space arguments. Consequently, IBP manipulations and field redefinitions in the bulk can leave non-trivial remnants on boundary correlators. The central problem addressed is determining when these boundary contributions are physically significant versus when they are mere redundancies, particularly in the context of massive particle exchanges (cosmological collider physics) and boost-breaking effective field theories (EFTs).

Methodology
The authors employ the Schwinger–Keldysh (in-in) formalism to compute cosmological correlators. Their approach is built on two main pillars:

1. Schwinger–Dyson Equations and Diagrammatic Reduction:
   The paper establishes a rigorous correspondence between local field redefinitions ($\Phi \to \tilde{\Phi} + f[\tilde{\Phi}]$) and boundary terms. By analyzing the invariance of the path integral under such redefinitions, the authors derive Schwinger–Dyson equations. These equations are translated into a set of four diagrammatic reduction rules (Section 2.1):

   • Rule 1 & 2: Relate boundary vertices involving conjugate momenta ($\Pi_0$) to variations in external and internal lines, effectively "cutting" lines and gluing operators to the boundary.
   • Rule 3 & 4: Relate bulk EoM vertices ($E_0$) to the vanishing of external lines or the collapse of internal propagators (generating contact diagrams).
   • Criteria for Survival: Based on these rules, the authors derive three criteria (Section 2.2) to determine if a boundary term yields a non-vanishing contribution in the late-time limit ($\eta_0 \to 0^-$):
     • Strict boundary contraction: Every conjugate momentum operator must be contracted with a boundary field via an equal-time commutator.
     • Absence of massive contraction: Remaining operators must not involve massive fields that decay rapidly at late times.
     • Absence of spatial derivatives: Boundary operators must not contain spatial derivatives, which introduce suppression factors of $\eta_0^2$.

2. Application to Specific Scenarios:

   • dS-Invariant Theories: The authors apply IBP and field redefinitions to correlators involving massive scalar exchanges ($\sigma$) and massless inflatons ($\phi$). They relate derivative couplings (e.g., $(\nabla \phi)^2 \sigma$) to non-derivative couplings ($\phi^2 \sigma$).
   • Boost-Breaking Theories: The analysis is extended to general EFTs of inflation where boost symmetry is broken (e.g., different sound speeds). They reduce general bulk interactions to a basis of independent shape functions and classify the surviving boundary contributions.
   • Bootstrap Verification: The results derived via IBP and field redefinitions are cross-checked against explicit computations using the cosmological bootstrap method (Appendix B), which solves differential equations derived from dS isometries.

Key Contributions and Results

• Establishment of Correspondence: The paper provides a systematic framework linking bulk field redefinitions to boundary terms. It demonstrates that field redefinitions generate both bulk EoM terms (which collapse propagators into contact interactions) and boundary terms.
• Classification of Boundary Effects:
  • In dS-invariant theories, the authors show that non-derivative exchange diagrams (e.g., $\phi^2 \sigma \times \phi^2 \sigma$) can be decomposed into derivative exchange diagrams (which are finite at late times) plus contact diagrams and specific boundary terms.
  • Crucially, they identify that late-time divergences (infrared divergences) in non-derivative exchange diagrams are entirely captured by the contact diagrams generated by the collapse of the internal propagator via the EoM operator. The accompanying boundary terms often cancel each other out (e.g., in the 4-point trispectrum) or decay, leaving the contact term as the sole source of the divergence.
  • For higher-derivative interactions, the authors derive recursion relations that connect exchange diagrams with different numbers of derivatives, reducing them to a basis of "box-type" vertices plus contact contributions.
• Boost-Breaking Classification: In the general EFT framework, the authors classify all possible boundary contributions to bispectra and trispectra. They identify specific combinations of boundary operators (e.g., $\phi' \phi \sigma \times \phi^2 \sigma'$) that survive the late-time limit, while others are suppressed by powers of $\eta_0$ or vanish due to branch-sum cancellations.
• Light Mass Regime: The paper addresses the regime where the intermediate mass $m < 3H/2$. In this case, late-time divergences cannot be fully resolved by standard IBP or field redefinitions alone. The authors show that bootstrap equations receive late-time decaying corrections, and the massless limit requires careful fixing of homogeneous solution coefficients to ensure consistency.

Significance and Claims
The paper claims to provide a more systematic understanding of boundary physics in inflationary spacetime, which has been overlooked in previous bootstrap analyses. Its primary significance lies in:

1. Clarifying Redundancies: It offers precise criteria for distinguishing between redundant bulk terms and physical boundary contributions, resolving ambiguities in previous studies where boundary terms were either ignored or assumed to be negligible.
2. Unifying Approaches: It demonstrates that the results obtained via the "bootstrap" (solving differential equations) and those obtained via "field theory" (IBP and field redefinitions) are consistent, with the latter providing a transparent diagrammatic interpretation of the former's analytic structure.
3. Systematic Reduction: By establishing recursion relations and a basis of independent shapes, the work simplifies the computational landscape for cosmological collider signals, particularly in boost-breaking scenarios where the number of independent templates can be large.
4. Future Directions: The authors suggest that this framework paves the way for constructing a Boundary Effective Field Theory (BEFT) of inflation to describe reheating processes and for exploring the connection between conformal anomalies on the boundary and field redefinitions in the context of de Sitter holography.

The work remains modest in its scope, focusing on tree-level massive exchanges and specific EFT setups, while leaving the investigation of non-local field redefinitions and loop-level effects for future research.