Multi-Matrix Quantum Mechanics, Collective Fields and Emergent Space

This paper investigates the quantum mechanics of bosonic multi-matrix Lagrangians, specifically focusing on three matrix models, to derive the effective Hamiltonian of the collective field and analyze its vacuum solution and stability.

The Gist

The Big Picture: Building Space from Scratch

Imagine you are trying to understand how a 3D world (like the room you are sitting in) is built. Usually, we assume space is just "there," like a stage where actors perform. But this paper asks a different question: What if space isn't a stage at all, but something that emerges from a bunch of tiny, interacting particles?

The authors are studying a specific type of math model called "Matrix Quantum Mechanics." Think of these models as a giant spreadsheet of numbers (matrices) that change over time. In these models, there is no pre-existing space. Instead, the "positions" of things are just numbers inside these spreadsheets. The goal of the paper is to show how a smooth, 3-dimensional space can pop out of this messy grid of numbers.

The Problem: Too Many Variables

The authors had previously studied a simpler version with just two matrices (two spreadsheets). They found a way to turn those two spreadsheets into a smooth 2D map of space.

However, our real universe has three spatial dimensions (up/down, left/right, forward/backward). To get a 3D universe, you need three matrices.

The problem with moving from two to three matrices is that it gets messy.

• The 2-Matrix Case: Imagine two groups of people. You can easily separate the "leaders" (diagonal numbers) from the "messengers" (off-diagonal numbers) connecting them.
• The 3-Matrix Case: Now you have three groups. The messengers from Group A talk to Group B, but they also talk to Group C, and Group B talks to Group C. It's like a chaotic party where everyone is shouting over everyone else. The math becomes incredibly complicated because these "messengers" are interacting with each other in a tangled web.

The Solution: The "Heavy" Filter

The authors found a clever trick to untangle this mess. They introduced a "mass" (a kind of weight) to the messengers.

The Analogy:
Imagine a crowded dance floor (the matrices).

• The leaders (diagonal numbers) are slow, heavy dancers who move gracefully. They represent the "space" we want to see.
• The messengers (off-diagonal numbers) are hyperactive, light dancers zipping around everywhere, connecting the leaders.

In the 3-matrix model, these hyperactive dancers are also bumping into each other, creating chaos. The authors' trick was to make these hyperactive dancers extremely heavy.

Because they are so heavy, they can't move fast or interact chaotically. They just sit there, vibrating slightly. This allows the authors to mathematically "integrate them out" (remove them from the active equation) and focus only on the slow, graceful leaders.

The Result: A New 3D Map

Once they removed the heavy, chaotic messengers, the remaining math for the leaders looked surprisingly clean. It transformed into a new kind of physics equation called a Collective Field Theory.

Instead of tracking thousands of individual numbers, the equation now describes a smooth, continuous "density" of space.

• The Shape: The authors solved this equation and found that the "space" that emerges isn't a perfect sphere. It's shaped like a squashed egg (an ellipsoid).
• The "Droplet": They call this a "droplet." It's a finite blob of space where the density of points is highest in the middle and fades to zero at the edges.
• The Twist: Because of the way they set up the math (choosing one matrix to be the "leader"), the space looks slightly different in one direction compared to the other two. It's like a balloon that is slightly stretched in one direction. The authors note this is likely just an artifact of their math setup, not a physical flaw in the universe.

Why This Matters (According to the Paper)

The paper claims this is a major step forward because:

1. It works: They proved that even with the messy interactions of three matrices, you can still derive a clean, 3D space if the "messengers" are heavy enough.
2. It scales: They showed that this method could theoretically be extended to 9 matrices (which is what the famous BFSS model for our universe uses). If you can handle 3, you can likely handle 9.
3. Stability: They checked if this new 3D space is stable. They found that if you wiggle the "droplet" slightly, it bounces back rather than falling apart. This suggests the emergent space is a solid, viable concept.

Summary

The paper is like a blueprint showing how to build a 3D house out of a pile of tangled wires.

• The Wires: The complex interactions between three matrices.
• The Trick: Making the tangled parts so heavy they stop moving, leaving only the structural frame.
• The House: A smooth, stable, 3D "droplet" of space that emerges naturally from the math.

The authors conclude that this method provides a practical way to understand how space itself might be an "emergent" property of quantum mechanics, rather than a fundamental building block of the universe.

Technical Summary

Technical Summary: Multi-Matrix Quantum Mechanics, Collective Fields and Emergent Space

Problem Statement
Matrix models, such as the BFSS and IKKT models, propose that spacetime emerges from the dynamics of $N \times N$ Hermitian matrices, where the fundamental degrees of freedom lack an ambient spacetime background. While the collective field method successfully describes emergent space in single-matrix models (yielding a 1D emergent space) and has been extended to two-matrix models (yielding a 2D emergent space), a rigorous analytical framework for multi-matrix systems ($p \geq 3$) remains elusive. The primary difficulty lies in the fact that diagonalizing one matrix leaves the off-diagonal components of the remaining matrices coupled. Unlike the single-matrix case, these off-diagonal modes cannot be ignored; they represent strings stretched between D0-branes. In previous attempts, working with gauge-invariant loop variables led to intractable Schwinger-Dyson equations, while naive diagonalization left interacting off-diagonal sectors that were difficult to integrate out analytically. The central problem addressed is whether a controlled analytical approach exists to derive an effective collective field theory for the diagonal degrees of freedom in a system with three or more interacting matrices.

Methodology
The authors employ a multi-step procedure to construct an effective collective field theory for a bosonic three-matrix model with a BMN-like mass deformation:

1. Model Selection: The study focuses on a $(0+1)$-dimensional $U(N)$ gauged matrix quantum mechanics with three Hermitian matrices ($X, Y, Z$). The action includes a kinetic term, a quartic commutator potential, and a mass deformation term involving a parameter $\nu$. This deformation stabilizes non-trivial vacua (fuzzy spheres) and ensures the off-diagonal modes are sufficiently massive.
2. Gauge Fixing and Splitting: The $U(N)$ gauge symmetry is fixed by diagonalizing one matrix, $X$. The remaining matrices, $Y$ and $Z$, are decomposed into diagonal parts ($y_i, z_i$) and off-diagonal parts ($y_{ij}, z_{ij}$). The diagonal variables represent the positions of $N$ D0-branes in an emergent 3D space, while off-diagonal variables represent string excitations.
3. Integrating Out Off-Diagonal Modes: The authors identify that in the presence of the mass deformation $\nu$, the off-diagonal modes are "heavy." They perform a Gaussian integration over these off-diagonal modes ($y_{ij}, z_{ij}$) in the Euclidean path integral. Crucially, unlike the two-matrix case where off-diagonal modes of the second matrix are independent oscillators, the three-matrix case features mixing between the off-diagonal sectors of $Y$ and $Z$ at the quadratic level. The authors treat this mixing via a matrix-valued kernel in the Gaussian approximation, expanding the resulting effective action in powers of $1/\nu$ to obtain a time-local effective potential.
4. Collective Field Transformation: The discrete diagonal eigenvalues $(x_i, y_i, z_i)$ are replaced by a continuous collective density field $\phi(\vec{x}, t)$ in the emergent three-dimensional space. The Vandermonde determinant arising from the gauge-fixing Jacobian is absorbed into the wavefunction, generating a cubic self-interaction term ($\phi^3$) in the Hamiltonian.
5. Vacuum Analysis: The authors derive the effective Hamiltonian for the collective field and solve for the large-$N$ vacuum solution (saddle point) by minimizing the potential energy subject to the normalization constraint. They subsequently analyze the stability of this solution by examining quadratic fluctuations around the saddle.

Key Contributions and Results

• Derivation of the 3D Collective Hamiltonian: The paper derives the explicit effective Hamiltonian for the emergent three-dimensional density field. The Hamiltonian includes:
  • A universal kinetic term involving the gradient of the conjugate momentum.
  • A cubic self-interaction term ($\sim \phi^3$) arising from the Vandermonde Jacobian.
  • A harmonic confinement term from the mass deformation.
  • A bi-local pairwise interaction term generated by integrating out the off-diagonal modes. Notably, this interaction kernel is anisotropic: the $x$-direction (associated with the diagonalized matrix) has a different coupling strength ($1/4\nu$) compared to the $y$ and $z$ directions ($1/8\nu$), a remnant of the gauge choice.
• Analytic Vacuum Solution: The authors find an analytic large-$N$ vacuum solution for the density profile $\phi_0(x, y, z)$. The solution takes the form of a "droplet" with a square-root profile:
  $$\phi_0 \propto \sqrt{\Lambda^2 - x^2 - \frac{1}{2}(y^2 + z^2)}$$
  The support of this distribution defines an ellipsoidal region in the emergent space, with the linear size scaling as $N^{1/8}$. The anisotropy in the ellipsoid shape reflects the gauge-fixed asymmetry between the diagonalized direction and the others.
• Stability Analysis: The paper demonstrates that the derived vacuum solution is a stable saddle point. By expanding the Hamiltonian to quadratic order in fluctuations, the authors show that the fluctuation spectrum contains no tachyonic directions (negative eigenvalues) in the leading large-$N$ approximation, provided the center of mass of the droplet is fixed.
• Generalization to $p > 3$: The authors outline how this construction generalizes to $p$ matrices (specifically targeting the $p=9$ bosonic sector of BFSS/BMN models). They argue that while off-diagonal mixing becomes more complex, the Gaussian approximation remains viable in the large-mass regime, leading to a similar collective Hamiltonian in higher dimensions.

Significance and Claims
The paper claims to provide the next necessary step in the program of deriving emergent spacetime dynamics directly from matrix quantum mechanics. Its primary significance lies in demonstrating that the collective field approach, previously limited to single and two-matrix models, can be extended to genuinely interacting multi-matrix systems where off-diagonal sectors mix.

The authors assert that their work provides evidence that the strategy of "gauge fixing $\to$ integrating out heavy off-diagonal modes $\to$ collective variables" is a viable route to emergent spacetime in interacting multi-matrix systems, at least within the approximation regime where off-diagonal modes are sufficiently massive. They emphasize that the three-matrix model captures the essential new ingredient (off-diagonal mixing) required for the general problem, suggesting that the construction can be systematically extended toward the bosonic sectors of BFSS and BMN models.

The paper remains modest regarding its scope, acknowledging that the derived theory is an effective description valid when the Gaussian approximation holds (large $\nu$). It notes that the description breaks down in the coincident-point limit and that a fully gauge-invariant formulation or the inclusion of fermions (for supersymmetric models) are necessary future steps to address the true ground state and cosmological applications. The work does not claim to have solved the full BFSS model but rather establishes a controlled framework for analyzing emergent geometry in a simplified, yet structurally representative, multi-matrix setting.