Ramp and plateau in bulk correlators within the disk topology in JT gravity

This paper demonstrates that the late-time ramp and plateau behavior of two-sided bulk correlators in Jackiw-Teitelboim gravity can be derived from the perturbative saddle expansion of the disk topology, revealing that this saturation occurs without requiring nonperturbative effects and that the dip-time scales inversely with black hole temperature.

The Gist

The Big Picture: The Black Hole Mystery

Imagine a black hole as a giant, cosmic trash compactor. In the 1970s, Stephen Hawking discovered that these trash compactors don't just sit there; they slowly leak out heat and radiation. Eventually, they vanish completely.

Here is the problem: In quantum mechanics (the rules of the very small), information cannot be destroyed. If you burn a book, the smoke and ash still contain the information about the book, even if it's scrambled. But if a black hole evaporates and disappears, where does the information about everything it swallowed go? This is the Black Hole Information Paradox.

For a long time, physicists thought the answer required "magic" (non-perturbative effects) or entirely new geometries of space-time to save the information. This paper suggests the answer might be much simpler: the information is already there, hiding in the subtle ripples of the black hole's gravity, waiting to be found.

The Setting: A Simple Toy Universe

The authors use a simplified model of gravity called JT Gravity. Think of this as a "training wheels" version of our universe.

• Real Universe: 3D space + 1D time (very complex).
• JT Gravity: 1D space + 1D time (like a flat sheet of paper).
• Why use it? It's simple enough to do the math exactly, but complex enough to still have black holes and the information paradox.

They are studying a specific setup: an Eternal Black Hole. Imagine a black hole that never dies, connected to a "mirror" black hole on the other side of a wormhole. These two sides are like two rooms connected by a secret tunnel.

The Experiment: Shouting Across the Room

To test if information is preserved, the authors imagine throwing a "message" (a massless scalar field, like a ripple in a pond) from one side of the black hole to the other. They measure how loud the message is when it arrives. This measurement is called a Correlator.

They look at two scenarios:

1. Same-Side Shout: You shout from the left room and listen in the left room.
2. Cross-Side Shout: You shout from the left room and listen in the right room (across the wormhole).

The Old Story: The Fading Echo

In the standard, "classical" view of black holes:

• Same-Side: The sound fades away quickly. This makes sense; the black hole absorbs the sound.
• Cross-Side: The sound also fades away exponentially. If you shout across the wormhole, the message dies out, and eventually, the two rooms become completely silent and disconnected.

If this were true, the information would be lost forever. The two sides would stop talking to each other.

The New Discovery: The "Dip-Ramp-Plateau"

The authors dug deeper. Instead of just looking at the main "loud" signal, they looked at the tiny, subtle ripples in the fabric of space-time (quantum fluctuations) that happen around the main signal.

They found a surprising pattern in the Cross-Side message:

1. The Dip (The Fade): First, the message fades away just like the old theory predicted. It hits a low point (the "dip").
2. The Ramp (The Rebound): Instead of staying silent, the signal starts to creep back up! It rises linearly, like a ramp.
3. The Plateau (The Constant Hum): Finally, the signal stops rising and settles at a steady, low level. It doesn't go back to being loud, but it never goes to zero.

The Analogy:
Imagine you are in a large, empty hall (the black hole). You clap your hands.

• Old Theory: The clap echoes, gets quieter, and then silence falls forever.
• New Theory: The clap echoes and gets quiet. But then, a faint, steady hum begins to rise from the walls. It's not a loud clap anymore, but it's a constant, unbreakable connection between the two sides of the hall. The information isn't lost; it's just been converted into a low-level hum.

Why This Matters: No Magic Required

The most exciting part of this paper is how they found this.

• Previous Belief: To get this "hum" (the plateau), you needed to invoke "non-perturbative" effects. In physics-speak, this means you had to assume the universe does something weird and magical that breaks the usual rules of math, or that the shape of space-time changes into something exotic (like a wormhole connecting the two sides in a new way).
• This Paper's Finding: You don't need magic. You don't need to change the shape of the universe. The "hum" appears naturally just by calculating the tiny, standard ripples (perturbations) around the normal black hole shape.

The Metaphor:
Think of a guitar string.

• Classical view: If you pluck it, it vibrates and then stops.
• This paper: If you listen really closely to the air around the string, you realize the vibrations never truly stop; they just become a tiny, constant vibration that keeps the string connected to the air. You didn't need to invent a new instrument to find this; you just needed to listen to the existing one more carefully.

The "Dip Time" and Temperature

The authors also found a relationship between the temperature of the black hole and when this "hum" starts.

• Hotter Black Hole: The "dip" happens faster, and the hum starts sooner.
• Colder Black Hole: The "dip" takes longer.
  This matches what we expect from the holographic principle (the idea that a black hole is like a hologram of a quantum system). It confirms that their math is consistent with the rest of physics.

Conclusion: What Does This Mean?

This paper suggests that the "Information Paradox" might not be a crisis requiring a revolution in physics. Instead, the information is preserved through subtle, standard quantum fluctuations that we just hadn't calculated correctly before.

The "Dip-Ramp-Plateau" structure is the fingerprint of a quantum system that never forgets. Even though the black hole looks like it's swallowing information, the math shows that the connection between the two sides of the universe never truly breaks. The information is safe, hidden in the quiet hum of the gravitational field.

In short: The universe is like a conversation that never truly ends; it just gets very quiet, but the connection remains.

Technical Summary

1. Problem Statement

The paper addresses the black hole information paradox, specifically focusing on the late-time behavior of correlation functions in Jackiw–Teitelboim (JT) gravity.

• The Conflict: Standard semiclassical calculations in eternal black hole geometries predict that two-sided correlators (connecting the two asymptotic boundaries of an eternal black hole) decay exponentially to zero at late times. This implies a loss of information and a violation of unitarity.
• The Expectation: In chaotic quantum systems (like the SYK model, which is dual to JT gravity), spectral observables and ensemble-averaged quantities exhibit a characteristic dip–ramp–plateau structure at late times. The "ramp" and "plateau" signify the persistence of correlations and the discrete nature of the energy spectrum.
• The Open Question: Previous literature suggested that reproducing the ramp and plateau in JT gravity requires non-perturbative effects (specifically, summing over different topologies like the "double trumpet" wormhole, which are exponentially suppressed by $e^{-S}$, where $S$ is the entropy). The authors ask: Is it possible to generate this late-time structure using only the perturbative expansion around the dominant disk topology, without invoking non-perturbative topological changes?

2. Methodology

The authors perform a rigorous perturbative analysis of bulk scalar field correlators in JT gravity, strictly adhering to the disk topology.

• Framework: They utilize the JT gravity action, which reduces to the Schwarzian action governing the boundary reparametrization modes ($f(t)$) after integrating out the bulk metric and dilaton.
• Setup:
  • They consider a massless scalar field $\phi$ in the eternal black hole background (Thermofield Double state).
  • They calculate the Hadamard function (symmetric two-point function) $H_{12} = \langle \{ \phi(x_1), \phi(x_2) \} \rangle$.
  • They distinguish between same-side correlators ($LL$, both points in one exterior) and two-sided correlators ($LR$, points in opposite exteriors).
• Perturbative Expansion:
  • The gravitational path integral is expanded around the classical saddle point (the disk geometry with $f(t) = t$).
  • They introduce fluctuations $\gamma(\tau)$ around the saddle: $f(\tau) = \tau + \lambda \gamma(\tau) + \dots$, where $\lambda^2 \propto G_N$ is the small gravitational coupling.
  • They compute the correlator up to next-to-leading order (NLO), i.e., order $\lambda^2$.
  • Correction to Previous Work: The authors explicitly correct a sign error found in a previous study (Ref [22]) regarding the correlator $\langle \gamma \dot{\gamma} \rangle$. This correction is crucial as it removes the unphysical linear growth in the same-side correlator and shifts the ramp behavior to the two-sided correlator.
• Calculation: They evaluate the Gaussian integrals over the fluctuation modes $\gamma$ and the Grassmann fields $\psi$ (arising from the measure on $\text{Diff}(S^1)/SL(2,\mathbb{R})$) to obtain the quantum corrections to the correlators.

3. Key Contributions

1. Perturbative Origin of the Ramp/Plateau: The paper demonstrates that the dip–ramp–plateau structure in two-sided correlators arises entirely from the perturbative expansion around the disk saddle. It does not require the inclusion of non-perturbative topologies (like the double trumpet) or $e^{-S}$ effects.
2. Resolution of the Sign Discrepancy: By correcting a sign error in the derivative of the two-point function used in prior literature, the authors reconcile the Schwarzian computation with the expectation that same-side correlators should decay, while two-sided correlators exhibit the ramp.
3. Distinction Between Power-Suppressed and Entropy-Suppressed: The authors clarify the hierarchy of contributions:
   • The plateau found here is power-suppressed ($\sim \lambda^2 \sim 1/N$).
   • Genuine non-perturbative effects (from other topologies) are entropy-suppressed ($\sim e^{-S} \sim e^{-N}$).
   • Therefore, the perturbative plateau dominates the late-time behavior in the regime where the expansion is valid.

4. Results

A. Same-Side Correlators ($LL$)

• Behavior: Both the tree-level ($\lambda^0$) and the first quantum correction ($\lambda^2$) exhibit exponential decay at late times.
• Result: $\langle H_{LL} \rangle \sim e^{-2\pi t_R / \beta}$.
• Implication: Correlations between points on the same side of the horizon are thermally suppressed, consistent with semiclassical expectations.

B. Two-Sided Correlators ($LR$)

• Tree-Level ($\lambda^0$): Exhibits pure exponential decay, similar to the $LL$ case.
• Next-to-Leading Order ($\lambda^2$): The quantum correction alters the late-time behavior significantly.
  • Instead of decaying to zero, the correction generates a linear ramp that drives the correlator toward a constant, non-zero value.
  • The full correlator $\langle H_{LR} \rangle \approx \langle \Gamma_0 \rangle + \lambda^2 \langle \Gamma_2 \rangle$ displays the complete dip–ramp–plateau structure.
  • The plateau value is negative and constant, scaling as $O(\lambda^2)$ (or $O(1/N)$).
• Dip Time ($t_{dip}$): The time at which the correlator reaches its minimum (the "dip") scales linearly with the inverse temperature $\beta$ (i.e., $t_{dip} \propto \beta$). This is consistent with holographic expectations for the scrambling time.

C. Boundary Correlators

• By taking the near-boundary limit ($z \to 0$), the authors show that the boundary CFT correlators inherit the same features.
• The one-sided boundary correlator decays exponentially.
• The cross-boundary boundary correlator exhibits the dip–ramp–plateau structure, with a plateau height dependent on $\beta$.

5. Significance and Implications

• Revisiting Semiclassical Gravity: The results challenge the naive view that semiclassical gravity (fixed background) inevitably leads to information loss. They show that controlled quantum fluctuations around a single saddle geometry are sufficient to encode the qualitative features of unitary evolution (the ramp and plateau).
• Role of Topology: While summing over topologies (wormholes) is often cited as the mechanism for the ramp in the SYK/JT correspondence, this paper proves that such topological changes are not strictly necessary to see the ramp; it emerges from the perturbative sector of the disk topology.
• Hierarchy of Scales: The work clarifies that the perturbative plateau ($1/N$) is parametrically larger than the non-perturbative plateau ($e^{-N}$). Thus, in the large-$N$ limit, the perturbative mechanism is the dominant source of late-time correlations.
• Future Directions: The authors note that the perturbative series is asymptotic (higher orders eventually lead to exponential growth), suggesting a need for resummation or non-perturbative completion at very late times. They propose direct tests in SYK models beyond leading order in $1/N$.

In summary, the paper provides a robust technical demonstration that the dip–ramp–plateau structure, a hallmark of quantum chaos and unitarity, is encoded in the perturbative fluctuations of JT gravity on the disk, offering a new perspective on how information is preserved in black hole evaporation without immediate recourse to non-perturbative topological transitions.