Proper-time functional renormalization in $O(N)$ scalar models coupled to gravity

This paper investigates the scaling solutions and critical properties of an $O(N)$ scalar field coupled to gravity in three and four dimensions using a proper-time functional renormalization group framework, confirming most qualitative and quantitative results from previous effective average action studies while highlighting specific differences in finite and large $N$ limits depending on the improvement scheme.

The Gist

The Big Picture: Mapping the Universe's "Zoom Levels"

Imagine you are looking at a digital photo of a forest. If you zoom out, you see the whole forest. If you zoom in, you see individual trees. If you zoom in even more, you see leaves, then veins in the leaves, then cells.

In physics, the universe works similarly. There are different "zoom levels" (called energy scales). At high energies (very small zoom), particles behave one way. At low energies (large zoom), they behave differently. The Renormalization Group (RG) is the mathematical tool physicists use to understand how the rules of the universe change as you zoom in and out.

This paper is about testing a specific, somewhat old-school map-making tool called the "Proper-Time" method to see if it works well for a universe that contains both matter (specifically, a group of particles called an O(N) scalar field) and gravity (the curvature of space-time).

The Two Competing Maps

The authors are comparing two different ways to draw this map:

1. The "Effective Average Action" (EAA) Map: This is the modern, popular GPS. It's been used for years and is known to be very accurate. The authors had used this map in previous studies.
2. The "Proper-Time" (PT) Map: This is an older, classic compass. It has some unique features, like being very good at respecting certain symmetries (rules that say the universe looks the same from different angles), but it's less commonly used for this specific job.

The Goal: The authors wanted to see if the "Proper-Time" compass gives the same results as the modern GPS when mapping the interaction between matter and gravity. They wanted to know: Does the old compass still work, or does it lead us astray?

The Experiment: Gravity and a Crowd of Particles

To test this, they set up a simulation of a universe with:

• Gravity: The fabric of space-time.
• A Crowd of Particles: Imagine $N$ different types of particles (like a crowd of people). They are "O(N) symmetric," which is a fancy way of saying they are all identical twins; swapping one for another doesn't change the physics.

They looked at this system in two different "worlds":

• 3 Dimensions: Like our everyday space (plus time).
• 4 Dimensions: The standard model of our universe (3 space + 1 time).

The "Fixed Points": The Universe's Anchor Points

As you zoom in and out, the rules of the universe usually keep changing. However, sometimes the rules hit a "sweet spot" where they stop changing. In physics, these are called Fixed Points.

Think of a Fixed Point like a gravitational anchor. No matter how much you zoom in or out, the physics at this specific point stays the same. These anchors are crucial because they tell us about the "universal behavior" of the universe—how things act regardless of the tiny details.

The authors were looking for two specific types of anchors:

1. The Gaussian Fixed Point: A simple, "boring" anchor where particles don't really interact with each other.
2. The Wilson-Fisher Fixed Point: A complex, "interesting" anchor where particles interact strongly. This is the kind of behavior seen in things like magnets or fluids near a boiling point.

The Results: A Tale of Two Schemes

The authors ran their simulations using two different settings for their "Proper-Time" compass, which they called Scheme C and Scheme B.

1. Scheme C (The "Unimproved" Compass)

• The Result: This version of the compass worked beautifully.
• The Analogy: It was like using a slightly older map that still had the right roads. The results matched the modern GPS (EAA) almost perfectly.
• The Finding: The "gravity-dressed" Wilson-Fisher anchor (the complex one) looked almost exactly like the one found in a universe without gravity. Gravity didn't mess things up much here. The critical properties (how the system behaves near the anchor) were very similar to what we expect from standard physics.

2. Scheme B (The "Improved" Compass)

• The Result: This version was more complicated and gave different answers.
• The Analogy: This was like using a map that had been "enhanced" with new data, but the enhancement changed the landscape.
• The Finding: In this scheme, gravity had a huge effect. The "Wilson-Fisher" anchor looked very different from the standard version. The rules of the game changed significantly.
  • In the standard version, there is usually one main "direction" where things can change (a relevant direction).
  • In this "Improved" scheme, they found three main directions where things could change.
  • The numbers describing how the system behaves (critical exponents) were quite different from the standard expectations.

The "Large Crowd" Limit ($N \to \infty$)

The authors also asked: "What happens if the crowd of particles becomes infinitely large?"

• The Result: When the crowd is huge, the two different compasses (Scheme C and Scheme B) agreed with each other completely.
• The Analogy: It's like a noisy party. If there are only a few people, the conversation depends on who is talking to whom (the specific scheme). But if there are thousands of people, the noise averages out, and everyone hears the same thing.
• The Finding: In this limit, gravity stopped affecting the matter particles' potential energy. The math became solvable exactly, and the results were clean and predictable.

The "Ghost" in the Machine (Imaginary Numbers)

One of the most interesting technical findings was about a specific number called $\omega$ (omega), which describes how fast the system returns to stability after a disturbance.

• In Scheme C, for small crowds (1 or 2 particles), this number became imaginary (involving the square root of -1). In physics, an imaginary number here often suggests the system is oscillating or behaving in a wobbly, unstable way.
• In Scheme B, the number stayed real, but the value was very different from the standard expectation.

Conclusion: Does the Old Compass Work?

The paper concludes that:

1. Yes, the Proper-Time method works. It confirms most of the pictures we saw with the modern GPS (EAA).
2. But, it depends on how you tune it. Depending on whether you use the "unimproved" (Scheme C) or "improved" (Scheme B) version of the Proper-Time regulator, you get different details about how gravity affects matter.
3. Gravity matters. Even though the "unimproved" scheme looked very similar to the gravity-free case, the "improved" scheme showed that gravity can drastically change the critical properties of the universe.

In short: The authors successfully tested an older mathematical tool against a modern one. They found that while the old tool generally agrees with the new one, the specific "settings" you choose can lead to very different predictions about how gravity and matter interact at the smallest scales of the universe.

Technical Summary

Technical Summary: Proper-time functional renormalization in O(N) scalar models coupled to gravity

Problem and Motivation
The paper addresses the non-perturbative renormalization group (RG) flow of an $O(N)$-invariant scalar field multiplet coupled to Einstein gravity in $d=3$ and $d=4$ dimensions. While the Functional Renormalization Group (FRG) based on the Effective Average Action (EAA) and the Wetterich-Morris equation has been extensively used to study asymptotic safety and critical phenomena in gravity-matter systems, this work focuses on the Wilsonian functional renormalization group framework employing a proper-time (PT) regulator.

The authors aim to test the utility of the PT flow equation, specifically Eq. (I.1), in searching for scaling solutions and determining critical properties. A primary motivation is to compare these results with previous findings obtained using the EAA framework (specifically Refs. [75, 89, 104]) to assess the robustness of the physical picture across different RG schemes. The PT approach is noted for its potential to preserve symmetries (crucial for gauge theories and gravity) and its ability to handle poles in even dimensions, acting as a variant of dimensional regularization.

Methodology
The study employs a truncation of the Wilsonian action $S_\Lambda$ that includes the Ricci scalar $R$ and a scalar potential, coupled non-minimally to gravity. The action is parameterized by two running functions, $F_\Lambda(\rho)$ (related to the non-minimal coupling and Newton's constant) and $U_\Lambda(\rho)$ (the effective potential), where $\rho = \frac{1}{2}\sum \phi_i^2$.

1. Flow Equations: The authors derive the RG flow equations for $F_\Lambda$ and $U_\Lambda$ using the proper-time flow equation (I.1) with a spectrally adjusted cutoff function $r(s, \Lambda^2 Z_\Lambda)$ defined in Eq. (I.2). This cutoff depends on a parameter $m$ and distinguishes between two schemes: Type B ($\epsilon=1$, "improved") and Type C ($\epsilon=0$, "unimproved").
2. Background Field Method: To handle the gravitational sector, the authors utilize the exponential split $g_{\mu\nu} = \bar{g}_{\mu\rho}(e^h)^\rho_\nu$ and a linear split for the scalar fields, matching the setup used in previous EAA studies to ensure a fair comparison.
3. Analytical and Numerical Approaches:
   • Analytical: The authors derive fixed-point solutions for the Gaussian limit and specific scaling solutions in $d=4$ and the large-$N$ limit ($N \to \infty$). Stability analysis is performed via linearization to extract critical exponents.
   • Numerical: For non-trivial fixed points (specifically the gravitationally dressed Wilson-Fisher fixed point in $d=3$), the authors employ two numerical techniques: the shooting method and the pseudospectral method. They implement an "improved" version of both, matching solutions at a fitting point and ensuring continuity of derivatives up to the third order to distinguish physical solutions from spurious ones.

Key Contributions and Results

• Comparison of Schemes (Type B vs. Type C):

  • Type C ($\epsilon=0$): The gravitational dressing of the Wilson-Fisher fixed point is minimal. The potential $u_*(x)$ closely resembles the pure Wilson-Fisher solution, and the critical exponent $\nu$ is very close to the classical value. However, the irrelevant exponent $\omega$ acquires an imaginary part for $N=1$ and $N=2$, a feature also observed in previous EAA studies.
  • Type B ($\epsilon=1$): The "improved" scheme yields significantly different results. The gravitational potential deviates strongly from the pure Wilson-Fisher case, and the critical exponent $\nu$ differs substantially from the classical value. This scheme reveals three non-trivial relevant directions for $N=1$, compared to one in the Type C scheme.

• Fixed Point Structure in $d=4$:

  • The space of fixed points includes the Gaussian fixed point and a matter-coupled Reuter-like fixed point (Eq. III.1).
  • A generalization where $f_*$ depends linearly on the field (Eq. III.3) exists only for the Type C cutoff.
  • The linearization around the Gaussian fixed point reveals two sets of critical exponents. The standard set (scalar field theory) is independent of $N$. A second set, exclusive to the gravity-matter coupling, depends on $N$ and the cutoff scheme. For $N < 16$, the number of relevant directions is determined by the integer part of $9/(16-N) + 1$.

• Large $N$ Limit:

  • In the limit $N \to \infty$, the distinction between Type B and Type C cutoffs vanishes. Gravity does not affect the fixed-point potential $u_*(x)$, which coincides with the flat spacetime result.
  • The critical exponents reproduce the well-known spectrum of the spherical model.
  • A unique non-trivial scaling solution exists for $d < 4$, characterized by a linear function $f_*(x)$, differing from the non-trivial $f_*(x)$ found in previous EAA analyses.

• Critical Exponents in $d=3$:

  • For finite $N$, the existence of the gravitationally dressed Wilson-Fisher fixed point depends on the scheme. In Type C, solutions exist for $N \lesssim 9$ and $N \gtrsim 8184$. In Type B, solutions exist for $N \lesssim 2$ and $N \gtrsim 10000$.
  • The critical exponent $\nu$ in Type C is close to the classical value, while in Type B it is significantly different.
  • The exponent $\omega$ becomes real for $N > 2$ in Type C, whereas it remains real but distant from classical values in Type B.

Significance and Claims
The paper claims that the proper-time functional RG framework, despite being an approximate scheme for the 1PI effective action, provides a consistent picture of critical phenomena in gravity-matter systems. The main conclusion is that most of the qualitative and quantitative picture previously uncovered using the EAA framework is confirmed in the PT framework.

However, the authors note specific differences:

1. The number of relevant directions and the values of critical exponents depend on the choice of the cutoff scheme (Type B vs. Type C) and the value of $N$.
2. The "improved" Type B scheme yields results that are more distinct from the classical Wilson-Fisher behavior compared to the Type C scheme.
3. The large-$N$ limit serves as a consistency check where scheme dependence disappears, and gravity decouples from the potential.

The work extends previous findings by exploring a broader dependence on $N$ and the regulator parameter $m$. The authors modestly state that while the results confirm the general structure of the UV-critical manifold, the presence of matter significantly modifies this structure, potentially allowing for different continuum limits. They identify the incorporation of a field-dependent wavefunction renormalization (full second-order derivative expansion) as a necessary next step for future work.