Preprint
Review

This version is not peer-reviewed.

A Finite Universe as a Laboratory for Quantum Gravity

A peer-reviewed version of this preprint was published in:
Universe 2026, 12(10), 303. https://doi.org/10.3390/universe12100303

Submitted:

31 August 2026

Posted:

01 September 2026

Read the latest preprint version here

Abstract
We explore the possibility that the observable Universe is finite and ask how this changes the prospects for testing quantum gravity. A finite overdense region embedded in a larger background is gravitationally unstable and can undergo global collapse, providing a natural prehistory for the observed expansion. If the collapse approaches a finite quantum-supported density \( \rho_G \), positive spatial curvature allows a nonsingular bounce to connect contraction to expansion. For sufficiently large total mass, the same finite region can naturally lie within its gravitational radius, while its global boundary may also contribute to the effective late-time dynamics. The central proposal of this review is that such a finite bouncing cosmology can act as a laboratory for quantum gravity. We use quadratic gravity and PT-symmetric quantum theory as concrete realizations, while emphasizing that the observational strategy is more general. The CMB scalar amplitude fixes the \( R^2 \) coupling \( \alpha \), large-angle parity provides an estimate of a primordial PT-breaking amplitude \( A_{\rm PT} \), and reheating may connect these quantities to the observed baryon asymmetry. The contracting phase provides a second source of structure, distinct from inflationary quantum fluctuations: nonlinear halos can become super-horizon before the bounce and later re-enter as relic compact objects. Their abundance and mass function depend on \( \rho_G \) and on the spectrum and curvature of the pre-bounce perturbation, allowing present observations of dark matter, early supermassive black holes, lensing and gravitational waves to constrain both the quantum endpoint of the collapse and its initial conditions. Strong-gravity observations provide an independent probe of the same microscopic completion through black-hole ringdown, extreme-mass-ratio inspirals, horizon absorption and tensor dynamics. The resulting framework is therefore over-constrained by observations across otherwise unrelated regimes. Relations among \( A_s \), \( \alpha \), \( A_{\rm PT} \), \( Y_B \), \( \rho_G \), spatial curvature, relic populations and strong-gravity observables provide a network of cross-consistency tests rather than isolated signatures. A finite Universe may thus preserve measurable information about both its pre-Big-Bang evolution and its high-density quantum state, turning otherwise inaccessible quantum-gravitational dynamics into a potentially falsifiable observational programme.
Keywords: 
;  ;  ;  ;  ;  ;  ;  ;  

1. Introduction

Understanding how quantum theory and gravity combine to describe the Universe remains one of the central challenges of modern physics [1,2]. General Relativity is expected to provide only the low-energy classical description of gravity, with quantum corrections becoming important at sufficiently high curvature or short distances [3,4]. Yet the regimes in which such effects become significant are usually assumed to lie far beyond direct experimental reach. This raises a fundamental question: can the Universe itself provide an arena in which quantum-gravitational physics becomes observable [5]?
In this work we explore the possibility that the answer is connected to a simple global property: the Universe may be finite. Finiteness changes the problem in an essential way. A finite gravitating system has a total mass, a characteristic size and global boundary conditions. Its evolution can therefore involve global gravitational instability in much the same way as smaller bound systems. Gravity is attractive, and the formation of structure throughout the observed Universe provides a continuous hierarchy of examples, from planets and stars to galaxies, clusters and superclusters. Extending this reasoning to the largest scales suggests that a finite cosmological region may itself originate as a collapsing overdensity embedded in a larger background.
From this perspective, cosmic expansion need not mark the absolute beginning of the Universe. It may instead represent the expanding branch of a gravitational system that previously underwent contraction. As the collapse proceeds, the density and curvature increase until the classical description ceases to be adequate. If quantum effects regulate the high-density phase and prevent a singularity, contraction can be connected to expansion through a nonsingular bounce. The Big Bang is then interpreted not as the creation of spacetime from an initial singularity, but as a transition between a contracting classical phase, a quantum-dominated state and the expanding Universe we observe.
The Black Hole Universe (BHU) provides a concrete realization of this idea. The essential ingredient, however, is not that the Universe is postulated to be a black hole, but that it is finite [6,7]. For a finite system of mass M, the gravitational radius is r s = 2 G M . Because the mean density required for an object of radius R to lie within its own gravitational radius scales as ρ ∝ R − 2 , the density threshold becomes progressively smaller for increasingly large systems. A cosmological region containing an enormous mass can therefore lie within its gravitational radius while having a very low mean density. In this sense, the black-hole character of the BHU follows naturally from the combination of finiteness, gravity and the large mass of the observable Universe.
Finiteness also opens several connections that are absent, or much less direct, in an infinite cosmological model. The global boundary may contribute to the effective late-time dynamics and provide a geometric interpretation of the cosmological constant [8]. A contracting phase allows structures to form before the bounce and potentially survive as relic compact objects in the expanding branch [9]. Positive spatial curvature provides a natural contribution to a nonsingular turning point. Most importantly for the present work, the high-density endpoint of gravitational collapse creates a physical regime in which a theory of quantum gravity can influence quantities that are subsequently observable.
This observation motivates the main theme of this review. We do not assume that one particular formulation of quantum gravity must provide the correct microscopic description. Instead, we ask how a finite bouncing cosmology can act as a laboratory for quantum gravity. A specific quantum-gravity model determines the dynamics near the high-density state, the spectrum of primordial fluctuations, the nature of the transition between contraction and expansion and possibly the production of matter and compact relics. The finite cosmological setting then propagates these microscopic ingredients into observables that can be measured at much later times.
Quadratic gravity provides a particularly useful example [10,11]. The addition of curvature-squared terms improves the ultraviolet behaviour of gravity, while the R 2 sector produces the scalaron dynamics underlying Starobinsky inflation. In this case the observed amplitude of primordial fluctuations fixes the dimensionless quadratic coupling α , and therefore fixes a characteristic energy and curvature scale of the inflationary phase. This raises the possibility that the high-density scale of the bounce is not an arbitrary parameter but can be related to quantities already measured in the cosmic microwave background. More generally, different quantum-gravity completions could predict different bounce scales, primordial spectra or strong-gravity signatures, allowing the finite Universe to discriminate between them.
A second example concerns the role of parity–time (PT) symmetry in the primordial quantum state. Quantum mechanics and relativistic quantum field theory possess fundamental discrete symmetries, while the observed cosmological state need not preserve them exactly. Large-angle CMB temperature fluctuations exhibit an intriguing preference for antisymmetric, odd-parity correlations. Within a PT-symmetric primordial framework, such observations can be interpreted as an estimate of the degree of PT symmetry breaking during the transition from the quantum state to the expanding semiclassical Universe. Direct-sum quantum theory provides one explicit realization of such a construction, but the observational interpretation considered here does not depend uniquely on that formulation.
This possibility becomes particularly interesting when the quantum-to-classical transition is connected to reheating and the origin of matter. In quadratic inflation, the scalar curvature evolves from the high-curvature quasi-de Sitter state toward the low-curvature radiation era, while the scalaron transfers gravitational energy into particles. Metric fluctuations are converted into matter and radiation perturbations during this transition. If the primordial state carries a measurable PT asymmetry, the same symmetry breaking may also participate in generating the observed matter–antimatter asymmetry, provided that baryon- or lepton-number-violating processes are active. This connects cosmological symmetry breaking, reheating, baryogenesis and leptogenesis within a common framework.
The observational implications extend well beyond the CMB. Compact structures formed during contraction may become super-horizon close to the bounce and survive into the expanding phase as relic black holes. Their abundance, mass spectrum and survival probability depend on the density ρ G of the high-density semiclassical state and can be constrained by the observed dark-matter density, the abundance of early supermassive black holes, lensing and gravitational-wave observations. The same quantum-gravity completion may also affect primordial tensor modes, black-hole quasinormal modes, horizon absorption and extreme-mass-ratio inspirals. Global finiteness can additionally leave signatures through spatial curvature, large-scale mode suppression, rotation or correlations between widely separated regions of the sky.
The central strength of this approach is therefore not any single observational anomaly or prediction. It is the possibility of testing the same underlying pre-Big-Bang physics through several independent observables. Quantities such as the primordial scalar amplitude A s , a PT-breaking amplitude A PT , the baryon asymmetry Y B , the bounce density ρ G , the abundance of relic compact objects, the population of early supermassive black holes and strong-field gravitational-wave observables may probe different aspects of the same high-curvature phase. A successful theory must account for these observations simultaneously.
This review develops that programme. We first examine the gravitational implications of a finite Universe and the possibility that the observed expansion follows from an earlier phase of collapse. We then discuss the high-density quantum state and the conditions required for a nonsingular bounce. Quadratic gravity and PT-symmetric quantum mechanics are introduced as concrete examples of microscopic frameworks that can be embedded in this cosmological setting. We subsequently explore inflation, reheating, primordial symmetry breaking, baryogenesis and leptogenesis, followed by the formation and survival of pre-bounce relics. Finally, we organize the resulting cosmological and strong-gravity signatures as a network of cross-consistency tests. The aim is not to advocate one unique theory of quantum gravity, but to show how a finite bouncing Universe can transform otherwise inaccessible quantum-gravitational dynamics into a potentially falsifiable observational programme.

2. A Finite Gravitating Universe

The starting point of the Black Hole Universe picture is not the existence of a black hole, but the assumption that the cosmological region we observe is finite. A finite gravitating system differs qualitatively from an idealized infinite FLRW Universe because it possesses a total mass, a characteristic size and a global boundary. These properties allow its evolution to be discussed not only locally, through the Friedmann equations, but also globally, as the dynamics of a self-gravitating region embedded in a larger spacetime.
A natural way to realize such a configuration is through a finite overdensity in an otherwise larger background.
Gravity is attractive, so sufficiently overdense regions tend to decouple from the surrounding expansion and collapse [12,13]. The same mechanism operates on familiar scales in the formation of bound structures such as stars and halos (see later Section 8.1 and Figure 7). The BHU proposal extends this logic to the largest possible scale: the cosmological region that later becomes our observable Universe may itself originate from the gravitational collapse of a finite perturbation in a parent background.
This does not require the parent spacetime to play an observational role in the subsequent evolution. Once the overdense region becomes causally isolated, its internal dynamics can be described independently. The essential point is that the observed expanding phase need not represent the absolute beginning of the system. It may instead correspond to the post-bounce evolution of a finite region that previously contracted.

2.1. Closed Geometry and Global Collapse

A homogeneous and isotropic finite region can be represented by a closed FLRW geometry [12,13],
d s 2 = − d τ 2 + a 2 ( τ ) d χ 2 1 − k χ 2 + χ 2 d Ω 2 , k > 0 .
The physical radius of a spherical boundary is
R ( τ ) = a ( τ ) χ ( τ ) ,
where χ need not be constant if the boundary is not comoving. Outside R > r s lies vacuum spacetime, which may be asymptotically Minkowski (i.e., Schwarzschild), deSitter, or Anti-deSitter [14]. This is illustrated in Figure 1. The matching details are not so important for the purposes of this review.
For a finite overdense region, the total enclosed mass M is finite [15]. The interior can therefore be matched to an exterior gravitational solution, and the associated gravitational radius is
r s = 2 G M c 2 .
where in the second step we have use c = 1 as in the FLRW metric above. The closed geometry is important because it provides the global contribution required for a turning point. The Friedmann equation may be written as
H 2 = 8 π G 3 ρ − k c 2 a 2 ,
where the positive spatial curvature term enters with the opposite sign to the matter density. During contraction the density generally increases, but if the high-density state approaches a finite value, the curvature term can balance the energy density at a non-zero scale factor and permit
H = 0 .
This is the basic geometrical condition required for a bounce.
The role of positive curvature should not be confused with finiteness itself. A compact space can in principle have zero or negative curvature. The physical statement relevant here is narrower: a finite overdense region produced by gravitational collapse naturally behaves as a closed system with positive curvature. In the Newtonian description of a spherical perturbation, this corresponds to a gravitationally bound system with negative total energy. Within this class of models, the curvature term provides the global contribution that allows contraction to reverse into expansion.

2.2. Why a Cosmological-Scale Region Can Become a Black Hole

The term “Black Hole Universe” can initially sound much more exotic than the underlying geometry actually requires. For a spherical region of radius R and mean density ρ ¯ , the enclosed mass is
M = 4 π 3 ρ ¯ R 3 .
The condition that the region lies within its own gravitational radius, R ≲ r s , gives
ρ ¯ BH ≃ 3 c 2 8 π G R 2 .
The important feature is the scaling
ρ ¯ BH ∝ R − 2 .
The larger the system, the smaller the mean density required for its gravitational radius to exceed its physical size. Stellar black holes are extremely dense because their radii are small. A cosmological black hole can instead have an enormous radius and a correspondingly low mean density.
The black-hole character of the BHU is therefore not introduced as an additional assumption about an exceptionally dense present-day Universe. It follows from the combination of finite total mass and sufficiently large size. In this sense, the central hypothesis is finiteness; the existence of a gravitational horizon is a consequence of the global mass distribution.
This also changes the interpretation of the horizon. In an ordinary black hole, one usually imagines an object embedded in an exterior spacetime and asks what happens to an observer falling inward. In the cosmological application, we occupy the interior region. The relevant question is therefore how the internal homogeneous geometry evolves when its finite boundary approaches or coincides with the gravitational radius.

2.3. The Global Boundary

A finite cosmological region requires a matching between its interior and exterior geometries. In the simplest BHU construction, the homogeneous interior is bounded by a junction surface at R, while the exterior is empty. The total mass inside the junction is conserved, and the corresponding gravitational radius r s provides a natural global scale.
For particular choices of the boundary motion, the junction can approach a null surface at
R = r s .
In this limit the boundary itself acts as a gravitational horizon. The resulting configuration combines a homogeneous cosmological interior with an exterior vacuum geometry while preserving the continuity conditions required by General Relativity.
This boundary is not merely a mathematical device. Because the Universe is finite, its global geometry can influence the effective dynamics seen by interior observers. In particular, the gravitational horizon introduces a fixed large-scale length that can be related to the asymptotic de Sitter scale.

2.4. Boundary Scale and the Cosmological Constant

A de Sitter spacetime with cosmological constant Λ possesses the characteristic horizon radius
R Λ = 3 Λ ,
in units where c = 1 . Restoring c where necessary, the corresponding expansion scale is
H Λ 2 = Λ c 2 3 .
In the finite BHU construction, the gravitational boundary can asymptotically approach this same scale,
r s = R Λ .
This provides a direct connection between the total mass of the finite region and the effective cosmological constant measured by interior observers.
The interpretation is conceptually different from introducing a uniform vacuum-energy density as an independent component of the cosmic fluid. The same Friedmann equation can still be written in the usual form,
H 2 = 8 π G 3 ρ m + ρ Λ − k c 2 a 2 ,
but the effective ρ Λ can instead encode information about the global boundary of the finite system. The local phenomenology may therefore resemble that of a cosmological constant even though its physical origin is associated with the finite gravitational geometry.
This interpretation is especially natural in a system that evolves from collapse to expansion. During the contracting phase the finite boundary defines the total gravitating mass. After the bounce, as the interior expands, the same boundary continues to constrain the global solution. At late times the interior can approach a quasi-de Sitter state set by the gravitational radius of the whole system. The early high-density phase and the late accelerated phase are then not unrelated additions to the cosmological model but different regimes of the same finite gravitational configuration.
The possibility that the observed value of the cosmological constant may encode quantum-gravitational information has a long history. Early approaches based on Euclidean quantum gravity attempted to derive a probability distribution for Λ . In particular, Elizalde and Gaztañaga showed that higher-order corrections to the effective action can qualitatively alter the strong preference for Λ = 0 obtained in Coleman’s original proposal, producing instead a smooth distribution around zero [16]. The present construction is physically different: Λ is associated with the global boundary of a finite gravitating region rather than selected from a probability distribution. Nevertheless, both approaches illustrate how the observed cosmological constant may be sensitive to physics beyond the classical Einstein action.

2.5. Finiteness as the Organizing Principle

The importance of finiteness can now be summarized. Once the cosmological region has a finite mass and size,
finiteness ⟶ global gravitational collapse , positive - curvature dynamics , a finite gravitational radius , a physical cos mological boundary .
These ingredients in turn open the possibility of a nonsingular bounce, an effective cosmological constant and the survival of pre-bounce structures. None of these possibilities requires that the microscopic physics at the high-density state be specified at this stage. They follow from the global structure of the finite gravitational system.
The role of quantum gravity enters at the next step. Classical collapse drives the density and curvature toward a regime where General Relativity can no longer be assumed to provide a complete description. The fate of the system then depends on the microscopic physics of the high-density state: whether the singularity is avoided, what fixes the maximum density ρ G , how contraction is connected to expansion and which structures can survive the transition. It is precisely here that a finite gravitating Universe becomes a laboratory for quantum gravity.

3. The High-Density Endpoint and the Bounce

A finite gravitating region undergoing collapse cannot remain in the classical regime indefinitely. As the physical size decreases, the density, characteristic momenta and curvature all increase. At sufficiently high density, quantum statistics, many-body correlations and eventually quantum-gravitational effects must become important. The central question is therefore what replaces the classical singularity once the collapsing system enters this regime.
The Black Hole Universe framework assumes that the collapse approaches a high-density semiclassical state with a finite characteristic density [7,9],
ρ ⟶ ρ G ,
rather than diverging. We use ρ G to denote the density of this quantum-supported state. Its precise microscopic origin is not fixed at this stage. It may arise from fermionic degeneracy and quantum exclusion, from higher-curvature gravitational dynamics, from a more fundamental quantum-gravity theory, or from a combination of these effects. The important assumption is that the high-density phase is regular and approximately stationary, so that the effective energy density remains finite.
For fermionic matter, there is a simple physical motivation for such behaviour. During collapse, the available phase-space volume decreases and the characteristic momenta increase. The quantum exclusion principle prevents identical fermions from occupying the same quantum state, driving the system toward an increasingly degenerate many-body configuration. The resulting quantum correlations and degeneracy pressure resist further classical compression. In the semiclassical description considered here, the limiting state approaches an approximately constant energy density.
Energy conservation in a homogeneous background gives
ρ ˙ = − 3 H ( ρ + p ) .
If ρ approaches a constant while H ≠ 0 , the effective pressure necessarily approaches
p ≃ − ρ G .
The high-density state therefore behaves gravitationally like a quasi-de Sitter phase. This should not be interpreted as saying that the individual microscopic particles possess negative pressure. Rather, p ≃ − ρ G is an effective description of the collective quantum state and its gravitational backreaction.

3.1. A Finite-Density Bounce

For a closed cosmological region, the Friedmann equation near the high-density state becomes
H 2 = H G 2 − k c 2 a 2 , H G 2 ≡ 8 π G 3 ρ G .
Because k > 0 , the curvature contribution can balance the finite density at a non-zero scale factor. The turning point occurs when
H = 0 ,
which gives
a b = k c H G .
The corresponding solution is
a ( τ ) = a b cosh H G ( τ − τ b ) ,
where τ b denotes the bounce time. The contracting branch,
τ < τ b , H < 0 ,
is continuously connected to the expanding branch,
τ > τ b , H > 0 ,
without ever reaching a = 0 or an infinite density. The classical singularity is therefore replaced by a smooth transition at finite scale factor and finite curvature.
The essential ingredients are particularly transparent:
finite high - density state + positive spatial curvature ⟶ nonsingular bounce .
Positive curvature alone does not prevent a singularity if the density diverges, while a finite density does not generically produce a turning point in a spatially flat Universe. In the BHU construction the two ingredients play complementary roles: quantum physics regulates the high-density state, while the finite closed geometry allows the gravitational evolution to reverse when the curvature radius R k = a b χ k = a b / k equals the Hubble radius r H = 1 / H G corresponding to ρ G .

3.2. The Characteristic Scale of the Quantum State

The density ρ G defines a natural Hubble scale,
H G = 8 π G ρ G 3 ,
and an associated causal scale,
R G ≡ c H G = 3 c 2 8 π G ρ G .
This scale is central to the observational programme developed later. It determines the shortest cosmological horizon reached during the high-density phase and therefore separates perturbations that remain causally connected through the bounce from those that become super-horizon.
Structures with characteristic physical scales larger than R G can lose causal contact with their surroundings during the approach to the bounce. Their evolution can then become effectively frozen until the expanding horizon grows sufficiently for them to re-enter. This opens the possibility that nonlinear structures formed during contraction survive the high-density transition and later appear as relic compact objects.
The value of ρ G is therefore not merely a microscopic parameter describing inaccessible physics. It determines an observable scale. The abundance and mass spectrum of relics, the survival of pre-bounce structures and potentially the spectrum of the largest primordial perturbations all depend on it. Conversely, these observations may be used to infer or constrain ρ G .
This reciprocal relation is one of the key reasons why a finite bouncing cosmology can act as a laboratory for quantum gravity.

3.3. From Classical Collapse to a Quantum State

The transition toward the bounce can also be viewed as a change in the nature of the physical description. Far from the high-density regime, matter follows approximately classical trajectories and can be represented by a fluid or a collisionless phase-space distribution. As contraction proceeds, the occupation of quantum states increases, interactions become stronger and exchange effects and entanglement become increasingly important.
In a path-integral description (see Figure 2):
A = ∫ D q exp i ℏ S [ q ] ,
classical behaviour emerges in the semiclassical limit because destructive interference suppresses contributions far from stationary-action histories [17,18]. Near the high-density state, however, the semiclassical approximation itself may cease to be adequate: the path integral need no longer be dominated by a narrow neighbourhood of a single classical trajectory, and a broader set of quantum configurations may contribute to the evolution [19].
The approach to the bounce can therefore be summarized schematically as
classical collapse ⟶ semiclassical regime ⟶ quantum - supported state .
This is the reverse of the transition usually emphasized in early-Universe cosmology. Standard inflationary calculations begin with quantum fluctuations and ask how they become classical perturbations. In a bouncing cosmology there is also a preceding process: initially classical structures and fields are driven by gravitational collapse toward an increasingly quantum state.

3.4. From the Quantum State Back to a Classical Universe

After the turning point, the process reverses. Expansion reduces the density, weakens interactions and increases the separation between the relevant degrees of freedom. The quantum high-density state evolves into perturbations of the expanding background, which subsequently decohere and become effectively classical.
The complete sequence is therefore
classical ⟶ quantum ⟶ classical .
The first transition is driven by collapse, increasing density and growing correlations. The second is driven by expansion, dilution and decoherence. The bounce sits between them as the regime in which the classical spacetime description is least adequate.
This viewpoint also changes the interpretation of the origin of primordial structure. Quantum fluctuations need not emerge from an otherwise structureless beginning. The quantum state at the bounce may contain information inherited from the contracting phase together with fluctuations generated by the high-curvature dynamics itself. During expansion these quantum fluctuations are converted into the classical matter and metric perturbations from which the observed cosmic web eventually develops.
The details of this conversion depend on the microscopic theory. A particular quantum-gravity model must specify the degrees of freedom present at the bounce, their quantum state, their characteristic scale and their subsequent evolution. It may also determine whether discrete symmetries such as PT are preserved or broken during the emergence of the expanding branch.

3.5. Deflation, Inflation and the Meaning of Curvature

The semiclassical bounce does more than replace the singularity by a smooth transition. If the high-density state approaches a quasi-de Sitter configuration with p ≃ − ρ G , then the same dynamics that produces accelerated expansion after the bounce also implies an approximately time-reversed accelerated contraction before it. The contracting branch can therefore contain a period of deflation, followed by an inflationary phase after the bounce.
This changes the usual interpretation of inflation [20]. In standard cosmology, inflation is often invoked to explain the observed smallness of spatial curvature because the curvature contribution to the Friedmann equation,
k c 2 a 2 ,
is rapidly diluted as the scale factor grows. Equivalently,
Ω k = − k c 2 a 2 H 2
is driven toward zero during a sufficiently long period of accelerated expansion.
In a bouncing cosmology, however, the pre-bounce evolution contains the opposite process. During deflation the scale factor decreases and the physical curvature contribution k / a 2 grows. To first approximation, a sufficiently symmetric inflationary phase after the bounce reverses the preceding deflationary evolution. The comoving curvature constant k itself is unchanged throughout the process: what changes is the physical curvature scale through its dependence on a.
The interpretation of the observed small curvature is therefore different. Post-bounce inflation need not be responsible for creating an exceptionally flat Universe from generic initial conditions. Instead, the combination of deflation and inflation can approximately restore the curvature scale inherited from the original finite perturbation. The present value of the curvature can then retain information about the initial conditions of the collapsing region.
For a finite overdensity embedded in a larger background, a small positive curvature corresponds to a perturbation that was very large in spatial extent but only weakly overdense. Such a perturbation can nevertheless become gravitationally bound and eventually collapse. In this sense, the observed near-flatness of the Universe need not imply that the underlying spatial curvature is exactly zero, nor that inflation has erased all memory of the initial state. A small residual curvature may instead be a fossil of the original perturbation from which the finite Universe formed.
This distinction is important because the inflationary phase is then assigned a different physical role. Its primary function is not to explain why the global curvature parameter is small, but to connect the high-density quantum state to the expanding semiclassical Universe and to generate the nearly scale-invariant spectrum of primordial fluctuations. The amount of residual curvature becomes an independent observable that probes the large-scale initial conditions of the collapse.

3.6. Two Origins of Primordial Structure

A bouncing cosmology also changes the interpretation of the origin of cosmic structure. In standard inflationary cosmology, essentially all primordial inhomogeneities are assumed to originate from quantum fluctuations generated during inflation and subsequently stretched beyond the Hubble radius [20]. These fluctuations are nearly Gaussian and nearly scale invariant, providing the initial conditions from which the observed large-scale structure develops.
In a finite Universe with a preceding contracting phase, there can be an additional source of structure [9]. Gravitational instability operates during contraction just as it does during expansion, and sufficiently overdense regions can collapse, virialize or form compact objects before the background reaches the high-density state. As the magnitude of the Hubble parameter grows during contraction, the Hubble radius
r H = c | H |
shrinks. Structures whose physical size becomes larger than r H therefore pass outside causal contact with the surrounding medium.
Once this occurs, their internal evolution becomes effectively decoupled from the large-scale background. These pre-existing nonlinear perturbations can survive the approach to the bounce as super-horizon relics and re-enter the causal horizon during the subsequent expansion. The expanding Universe can therefore contain two physically distinct contributions to its primordial density field,
δ ( x ) = δ inf ( x ) + δ relic ( x ) ,
where δ inf denotes the approximately Gaussian, nearly scale-invariant fluctuations generated during inflation, while δ relic represents nonlinear structures inherited from the contracting phase.
The two components need not have the same statistical properties or physical origin. Inflationary fluctuations naturally dominate the smooth primordial spectrum measured in the CMB and the large-scale matter distribution. Relic perturbations can instead be rare, highly nonlinear and strongly non-Gaussian, contributing primarily on the scales associated with compact objects, early halos or black-hole seeds.
This separation is particularly relevant for the formation of primordial compact objects. In conventional scenarios, producing a significant abundance of primordial black holes from the same nearly Gaussian fluctuation field measured in the CMB generally requires access to the extreme tail of the distribution, or a substantial enhancement and often non-Gaussian modification of the primordial spectrum on smaller scales. In the present framework this requirement is relaxed because the compact-object seeds need not originate from exceptionally large inflationary fluctuations.
Virialized structures formed during contraction can already possess overdensities far above the threshold required for black-hole collapse when they later re-enter the horizon. Their existence is therefore compatible with an inflationary fluctuation field that remains close to Gaussian and nearly scale invariant. The two observational requirements,
nearly Gaussian primordial fluctuations & strongly nonlinear compact seeds ,
can be satisfied by two different populations rather than by forcing a single primordial spectrum to account for both.
This provides a natural way of combining the conventional successes of inflationary structure formation with genuinely pre-bounce relics. The inflationary component seeds the large-scale cosmic web, while the relic component may contribute to dark matter, provide early seeds for supermassive black holes and generate distinctive gravitational-wave or lensing signatures. Their relative abundance and characteristic mass scale depend on the high-density parameter ρ G and on the duration of the contracting and inflationary phases, providing another route by which observations can constrain the physics of the bounce.

3.7. A Bounce Without Committing to a Unique Quantum Theory

It is useful to keep the logical ingredients distinct. The finite BHU geometry supplies the global conditions for collapse and positive curvature. The assumption of a finite-density quantum state replaces the singular endpoint. Together these ingredients allow a nonsingular bounce. They do not by themselves identify the microscopic theory responsible for that state.
This distinction allows the BHU to be used as a framework in which different theories of quantum gravity can be tested. A candidate theory should ideally predict quantities such as
ρ G , R G , the duration and dynamics of the high - density phase ,
together with the spectrum and symmetry properties of the perturbations emerging from it.
These predictions can then be confronted with independent observations. In the following sections we use quadratic gravity as a concrete example because its R 2 sector provides a well-defined high-curvature inflationary dynamics whose normalization is already constrained by the CMB. We will also consider PT-symmetric quantum descriptions as possible realizations of the quantum state across the bounce. Neither construction is required by the finite-universe argument itself. Their value is that they turn the otherwise unspecified high-density state into a predictive physical model.

4. Quantum Gravity in a Finite Bouncing Cosmology

The finite-universe construction described above does not by itself determine the microscopic physics of the bounce. General Relativity provides the global gravitational evolution through most of the contracting and expanding phases, but as the density and curvature approach the scale ρ G , the classical description must eventually be replaced by a quantum one. The role of a quantum-gravity completion is therefore not to create the finite cosmological setting, but to determine what happens at its high-density endpoint.
This distinction is useful because it separates two logically different questions. The first is global and essentially classical: can a finite gravitating region collapse, reach a minimum size and subsequently re-expand? The second is microscopic: what quantum dynamics regulates the high-curvature state, prevents a singularity and determines the perturbations and particle content emerging from the bounce?
A successful quantum-gravity completion of a finite bouncing cosmology should ideally address several requirements simultaneously. It should recover General Relativity at low curvature, provide a consistent quantum description at high curvature, preserve unitarity, regulate or remove classical singularities and produce a semiclassical expanding state compatible with observations. In addition, if the theory is to be empirically useful, it should determine at least some of the characteristic scales and symmetry properties of the bounce rather than leaving them as arbitrary initial conditions.
In the present framework, particularly important quantities include
ρ G , H G , R G = c H G ,
together with the duration of the high-density phase, the primordial fluctuation spectrum and the symmetry properties of the quantum state. These quantities control very different observables: the amplitude and scale dependence of CMB perturbations, the survival of pre-bounce relics, the minimum horizon scale, possible matter–antimatter asymmetries and strong-gravity signatures.
The finite bouncing Universe therefore provides an unusual setting in which a quantum-gravity theory can potentially be tested indirectly over a very wide range of scales. The microscopic dynamics near the bounce may influence observables that later appear in the CMB, large-scale structure, relic compact objects or gravitational waves. Different quantum-gravity completions need not make the same predictions for these quantities, allowing the cosmological setting itself to act as a discriminator between them.

4.1. What Should a Quantum-Gravity Completion Explain?

At the most basic level, the classical singularity must be replaced by a physically admissible quantum state. This does not necessarily require that curvature vanish or that the bounce be completely classical. It requires instead that physical observables remain finite and that the evolution connect the contracting and expanding semiclassical phases without loss of probability or predictive power.
A useful way of organizing the required physics is
classical collapse ⟶ quantum high - density phase ⟶ classical expansion .
The quantum theory should then answer four related questions.
First, what sets the limiting density ρ G ? In a phenomenological bounce this quantity can simply be introduced as a new scale. A more predictive theory should relate it to fundamental couplings or dynamically generated scales.
Second, what determines the duration and geometry of the bounce? This controls the minimum causal radius and therefore which structures remain sub-horizon and which become frozen relic perturbations.
Third, what quantum state emerges into the expanding branch? The answer determines the primordial scalar and tensor perturbations, their parity properties and their correlations.
Finally, how does the quantum state convert into ordinary matter and radiation? The reheating or quantum-to-classical transition is where the gravitational degrees of freedom become the particle content of the hot Big Bang, and it may also be where primordial symmetry breaking is transferred to matter.
These questions are closely related but need not have the same microscopic origin. Matter degeneracy can regulate compression, higher-curvature terms can dominate the gravitational action, discrete spacetime symmetries can determine the structure of the quantum state and particle interactions can control reheating. A complete theory should explain how these ingredients fit together.

4.2. Quadratic Gravity as a Worked Example

Quadratic gravity provides a particularly useful example because it extends General Relativity by the leading local curvature-squared operators and becomes increasingly relevant in the high-curvature regime [21,22]. In four dimensions the action may be written as
S QG = ∫ d 4 x − g M Pl 2 2 R + α 2 R 2 + β 2 W μ ν ρ σ W μ ν ρ σ ,
where M Pl is the reduced Planck mass, W μ ν ρ σ is the Weyl tensor and α and β are dimensionless couplings.
Interestingly, the sensitivity of cosmological predictions to higher-order terms in the gravitational effective action was already apparent in early studies of the quantum probability distribution of the cosmological constant [16].
The importance of the quadratic terms is twofold. First, they improve the ultraviolet behaviour of the gravitational propagator. Schematically,
D ( k ) ∼ 1 k 2 ⟶ D ( k ) ∼ 1 k 4 , k 2 → ∞ ,
making the theory perturbatively renormalizable. Second, the additional curvature terms introduce new dynamical sectors that become relevant precisely where classical General Relativity predicts large or divergent curvature.
The R 2 term introduces a scalar degree of freedom, the scalaron, whose characteristic mass is
M 2 ≃ M Pl 2 6 α .
In a homogeneous and isotropic FLRW background the Weyl tensor vanishes,
W μ ν ρ σ = 0 ,
so the background high-curvature dynamics is governed primarily by the Ricci and R 2 sectors. The same scalaron dynamics that regulates the high-curvature evolution naturally generates the quasi-de Sitter phase of Starobinsky inflation.
Quadratic gravity is therefore particularly well suited to the finite-bounce picture. The contracting Universe can approach a high-curvature quasi-de Sitter state, pass through a nonsingular transition and emerge into an inflationary expanding phase. To first approximation, the high-density evolution possesses the qualitative structure already anticipated from the semiclassical bounce: deflation during contraction followed by inflation during expansion.
The significance of this example is that the relevant inflationary scale is not arbitrary. The observed amplitude of primordial scalar perturbations fixes α to be of order
α ∼ 10 9 ,
as discussed in detail below. Consequently, the same CMB measurement that determines the amplitude of primordial structure also fixes the characteristic scalaron mass and high-curvature scale of the theory.
This raises an important possibility for the BHU framework: the density ρ G of the semiclassical state may ultimately be related to the inflationary scale determined by α . If this identification can be established, the bounce density would no longer be an independent parameter. It could instead be predicted by a coupling already measured through the primordial fluctuation spectrum.
At present this relation should be regarded as a hypothesis rather than an established result. The high-density state of the BHU has been motivated from many-body quantum degeneracy, whereas the quasi-de Sitter state of quadratic gravity follows from curvature dynamics. Demonstrating that the two descriptions correspond to the same physical state requires a microscopic treatment of matter and gravity together. Nevertheless, their shared properties—finite density, quasi-de Sitter evolution and a natural transition between contraction and expansion—make the comparison particularly suggestive.

4.3. The Role of the Weyl Sector

The Weyl-squared term does not affect an exactly homogeneous FLRW background, but it becomes important for departures from homogeneity and isotropy. It therefore controls aspects of the physics that are directly relevant to perturbations, tensor modes and strong-gravity environments.
In conventional quadratic gravity the additional spin–2 degree of freedom associated with the Weyl term leads to the well-known ghost problem. Recent formulations [11] have proposed alternative quantizations in which this sector is not interpreted as a propagating negative-norm particle. Direct-sum quantization provides one explicit example, in which the additional spin–2 mode is described through complementary PT-related sectors and does not belong to the physical asymptotic particle spectrum.
For the present review, the detailed implementation is secondary. The important point is that any viable quadratic quantum-gravity model must preserve the ultraviolet improvement provided by the Weyl sector while maintaining a positive and unitary physical state space. Different resolutions of the spin–2 problem may lead to different observable consequences for tensor perturbations, black-hole ringdown or horizon dynamics.
This is precisely where the finite-universe approach becomes useful. The microscopic treatment of the Weyl sector is not merely a formal issue at inaccessible energies: it can potentially influence primordial gravitational waves, parity-dependent tensor modes, quasinormal modes and extreme-mass-ratio inspirals.

4.4. Singularity Avoidance and the Finite-Action Criterion

Quadratic curvature terms also modify the status of singular gravitational histories. Near a classical singularity, quantities such as R 2 or W 2 can grow rapidly, causing the gravitational action to diverge. In a path-integral description,
A ∼ ∫ D g exp i ℏ S [ g ] ,
histories with arbitrarily large action acquire rapidly oscillating phases and may be strongly suppressed relative to smoother configurations.
This motivates the finite-action criterion: physically relevant histories should possess a finite gravitational action. In higher-curvature gravity, this requirement can disfavor classes of singular and strongly anisotropic solutions and favour regular high-curvature configurations.
The finite-action argument alone does not uniquely determine the bounce. It constrains which histories can contribute significantly to the quantum evolution, while the global closed geometry and the matter state determine whether contraction actually reverses into expansion. In the BHU construction, these ingredients are complementary:
finite geometry → positive - curvature turning point , high - density quantum state → finite ρ G , quantum gravity → regular high - curvature dynamics .
A successful microscopic theory should ultimately explain all three within a single framework.

4.5. From Microscopic Theory to Observable Parameters

The most important reason for embedding quantum gravity in a finite cosmology is that the fundamental parameters of the microscopic theory can become connected to measurable quantities.
For quadratic gravity, the schematic chain is
A s ⟶ α ⟶ M ⟶ high - curvature dynamics .
If the same dynamics determines ρ G , this extends to
A s ⟶ α ⟶ ρ G ⟶ R G ⟶ relic survival and abundance .
The same microscopic theory can independently affect tensor perturbations and strong-gravity observables through its additional gravitational degrees of freedom. Thus a single quantum-gravity model may be constrained simultaneously by the CMB, the bounce scale, the relic population and black-hole gravitational-wave observations.
This is the central methodological point of the present paper. Quadratic gravity is useful not because it must be the unique theory describing the bounce, but because it illustrates how a concrete quantum-gravity completion can transform an otherwise phenomenological high-density state into a network of quantitative predictions. Other microscopic theories can be embedded in the same finite cosmological setting and confronted with the same observational programme.
The next step is to examine a more general property that may be required of the primordial quantum state: its behaviour under parity and time reversal. There are several independent reasons to take this symmetry seriously before considering how it may be broken. In quadratic gravity, the additional spin–2 sector that appears as a ghost in conventional quantization suggests that the usual single-sector description may be incomplete, while PT-related formulations provide one possible route to a unitary interpretation. In black-hole physics, the information-loss problem and the global structure of Einstein–Rosen geometries similarly point toward complementary descriptions associated with different causal regions or time orientations. Related ideas have appeared in different forms in the work of Einstein and Rosen, Schrödinger, ’t Hooft and others, long before their recent formulation in direct-sum quantum theory [23]. Strong-gravity observations may provide an additional test: the approximate isospectrality of axial and polar black-hole quasinormal modes reflects a deep symmetry of the underlying perturbation equations, and departures from this structure could reveal how such symmetries are modified by quantum gravity.
These considerations motivate treating PT symmetry not as a phenomenological assumption introduced to explain a particular observation, but as a possible organizing principle of the underlying quantum theory. Direct-sum quantization provides one explicit realization, but it is not unique and will be used here primarily as a worked example. If the fundamental high-density state is PT symmetric while the observed expanding semiclassical branch selects one orientation, PT symmetry must then be broken during the quantum-to-classical transition. Such symmetry breaking can leave measurable imprints in the primordial perturbations and, through CPT invariance, may also be connected to the particle–antiparticle asymmetry of the Universe.

5. PT Symmetry and the Primordial Quantum State

The discussion so far has focused on the geometry and dynamics of a finite bouncing Universe. We now turn to a different question: what symmetry should characterize the quantum state that connects the contracting and expanding branches?
A natural possibility is that the fundamental quantum description is invariant under the combined action of parity and time reversal, PT, even if the semiclassical Universe observed on one side of the bounce is not. In this picture, PT symmetry belongs to the underlying quantum state or evolution law, while the emergence of a particular expanding branch corresponds to a spontaneous or effective breaking of that symmetry.
This distinction is important. If PT asymmetry were introduced only at the phenomenological level, it would amount to adding a new ingredient in order to explain an observational anomaly. If PT symmetry instead follows from deeper requirements of quantum gravity, such as unitarity, complementarity or the consistent treatment of apparently ghost-like sectors [11], then any observed PT breaking becomes a probe of how the fundamental quantum state gives rise to the classical Universe.

5.1. PT Symmetry as an Organizing Principle

In ordinary quantum mechanics, parity reverses spatial orientation,
P : x → − x ,
while time reversal reverses the direction of time,
T : t → − t ,
and acts antiunitarily on the quantum state. Their combined action relates configurations with opposite spatial and temporal orientation.
The relevance of PT symmetry extends beyond conventional Hermitian quantum mechanics. A Hamiltonian need not be Hermitian in the conventional sense to possess a real spectrum and unitary evolution if it satisfies appropriate PT-symmetry conditions and the physical inner product is defined consistently. More broadly, PT-related sectors can provide complementary descriptions of a single quantum system.
For quantum gravity this possibility is especially suggestive. Horizons divide spacetime into regions that need not admit a single global time orientation accessible to one observer. Black-hole interiors, Einstein–Rosen bridges and cosmological bounces naturally introduce pairs of causally or temporally complementary regions. A quantum description that treats one orientation as fundamental while discarding the other can therefore obscure symmetries that are present in the full geometry.
The idea that a complete quantum description may require complementary sectors has appeared in several forms. Einstein and Rosen introduced bridge geometries connecting apparently distinct spacetime regions; Schrödinger emphasized the nonlocal and relational character of entangled quantum states; and more recent approaches, including antipodal constructions and horizon-based descriptions, have stressed that information associated with different causal regions may belong to a single global quantum state. Direct-sum quantum theory provides one explicit modern realization in which opposite PT orientations are incorporated from the outset [23].
The common motivation is not that all these constructions are equivalent, but that quantum gravity may require a description more global than the semiclassical viewpoint available to a single observer.

5.2. Direct-Sum Quantization as One Realization

In direct-sum quantum theory, the Hilbert space is decomposed into two complementary sectors,
H = H A ⊕ H B ,
which are related by PT transformation. The quantum state can be represented schematically as
Ψ = 1 2 ψ A ψ B ,
with the two components evolving with opposite microscopic time orientations.
The corresponding evolution operator takes a block-diagonal form,
U ( τ ) = e − i H ^ A τ / ℏ 0 0 e + i H ^ B τ / ℏ ,
so that the complete state evolves unitarily even though the two sectors carry opposite time orientation.
Figure 3. The picture depicts the new understanding of quantum harmonic oscillator in a direct-sum Hilbert space. Time is a parameter in quantum theory. In contrast, the spatial position is an operator. A quantum state here is described by a direct-sum of two components in parity conjugate points in physical space.
Figure 3. The picture depicts the new understanding of quantum harmonic oscillator in a direct-sum Hilbert space. Time is a parameter in quantum theory. In contrast, the spatial position is an operator. A quantum state here is described by a direct-sum of two components in parity conjugate points in physical space.
Preprints 231015 g003
In this framework, the contracting and expanding sides of a bounce can be interpreted as semiclassical manifestations of the two complementary sectors. The full quantum state can preserve PT symmetry even though an observer inhabiting one branch experiences a definite arrow of time.
This construction is useful because it illustrates explicitly how a PT-symmetric fundamental state can coexist with a PT-asymmetric observable Universe. It should not, however, be regarded as the unique realization of this idea. The observational arguments developed below require only that the primordial quantum state possess an approximate PT symmetry and that the emergence of the expanding branch permits this symmetry to be broken.

5.3. Symmetry Breaking at the Quantum–Classical Transition

Suppose that the high-density state at the bounce is PT symmetric. The two semiclassical branches can then be represented schematically as
PT | expanding 〉 = | contracting 〉 .
The full quantum state may contain both orientations symmetrically, while the classical Universe accessible to us corresponds to one branch.
The emergence of a definite branch therefore resembles spontaneous symmetry breaking. The underlying equations and quantum state can preserve PT, but the realized semiclassical history does not.
We may parameterize the departure from exact PT symmetry by a dimensionless amplitude,
A PT = 0 for exact PT symmetry ,
with non-zero A PT characterizing the asymmetry of the selected branch.
The precise microscopic definition of A PT depends on the theory. It may correspond to an imbalance between PT-related sectors, a relative phase, an asymmetry in the primordial two-point function or a difference in the occupation of parity-related modes. For the observational discussion it is useful to treat A PT as an effective parameter characterizing the amount of primordial PT breaking that is transferred to cosmological perturbations.

5.4. Parity in the Primordial Perturbations

The cosmic microwave background (CMB) provides a particularly direct probe of parity on the largest observable scales (see Figure 4). Expanding the temperature anisotropy in spherical harmonics,
Δ T T ( n ^ ) = ∑ ℓ m a ℓ m Y ℓ m ( n ^ ) ,
parity acts as
Y ℓ m ( − n ^ ) = ( − 1 ) ℓ Y ℓ m ( n ^ ) .
Even multipoles are therefore parity symmetric, whereas odd multipoles are parity antisymmetric.
In standard inflation, the primordial spectrum is assumed to be parity symmetric, so the ensemble prediction contains no systematic preference for even or odd multipoles. Direct-sum inflation (DSI), by contrast, predicts a scale-dependent parity asymmetry on the largest scales as a consequence of the PT structure of the primordial quantum state. The effect is not fitted phenomenologically to the observed CMB: the parity modulation follows from the theory without introducing an additional free parameter [25].
A convenient statistic is the ratio of even- to odd-parity power,
R T T = C + C − ,
where C + and C − denote suitably averaged even- and odd-multipole powers. Exact parity balance corresponds statistically to
R T T ≃ 1 .
Planck temperature maps instead show a persistent preference for odd parity on the largest angular scales, corresponding to an asymmetry of order 20 % . The DSI prediction lies close to the observed value, whereas the standard inflationary prediction is centered on parity balance.
The statistical interpretation of this result is important. Large-scale CMB anomalies are often described as being limited by cosmic variance because only a small number of independent modes are available at low multipoles. This is correct when asking how unusual a particular realization is within a fixed cosmological model, but it is not the same question as comparing two models that predict different probability distributions for the same observables.
In Bayesian model comparison one considers
p ( M 1 | D ) p ( M 0 | D ) = p ( D | M 1 ) p ( D | M 0 ) p ( M 1 ) p ( M 0 ) ,
where M 0 and M 1 denote the competing primordial models. Cosmic variance is already incorporated in the likelihoods p ( D | M i ) predicted by each model. It broadens the expected distributions, but it does not erase the information contained in the fact that the observed realization may lie systematically closer to one distribution than to the other.
Using 10 6 Monte Carlo realizations of the Planck temperature sky, the low-multipole power spectrum gives a cumulative Bayes factor of approximately
B DSI , SI ∼ 150
by ℓ ≃ 20 , while the quadrupole alone gives B ∼ 32 . More significantly, the joint distribution of the quadrupole amplitude C 2 , the even-to-odd parity ratio R T T and the antipodal correlation C ( π ) gives
B DSI , SI ≃ 650 ,
in favour of the parity-asymmetric DSI prediction over standard inflation.
The key point is therefore not that cosmic variance is absent, but that it is explicitly included in the comparison. Cosmic variance limits how precisely the underlying low-multipole spectrum can be reconstructed from a single sky, but it does not prevent Bayesian discrimination between models that predict different statistical distributions for that sky. For the combined parity-sensitive statistics considered here, the observed CMB is several hundred times more probable under the DSI prediction than under standard parity-symmetric inflation.
The angular correlation function provides an equivalent real-space description,
C ( θ ) = Δ T T ( n ^ ) Δ T T ( n ^ ′ ) , n ^ · n ^ ′ = cos θ .
At antipodal separation,
θ = π ,
even and odd multipoles contribute with opposite signs. The observed suppression and negative character of the large-angle antipodal correlation are therefore closely connected to the same odd-parity preference and contribute independently to the Bayesian evidence.

5.5. From CMB Parity to a PT-Breaking Amplitude

The observed parity imbalance can be expressed phenomenologically as an effective amplitude of primordial PT breaking. Schematically,
A PT ∼ P odd − P even P odd + P even ,
where the precise numerical definition depends on the multipole weighting and estimator adopted.
Figure 5. Joint distribution of three statistics characterizing the large-scale CMB parity asymmetry: the quadrupole amplitude C 2 , the temperature parity ratio R T T , and antipodal correlation C ( π ) . Grey contours and one-dimensional distributions show the probability density obtained from Monte Carlo realizations of the observed sky. Red markers indicate the mean predictions of the SI model, while blue markers show the DSI prediction. The coloured dashed lines correspond to measurements from the four Planck component-separated maps (Commander, SMICA, SEVEM and NILC [26]), illustrating that the differences between foreground-cleaning methods are much smaller than the intrinsic cosmic variance. The observed realization lies much closer to the DSI prediction than to the SI expectation, giving a join Bayes factor B = 650 in favour of DSI.
Figure 5. Joint distribution of three statistics characterizing the large-scale CMB parity asymmetry: the quadrupole amplitude C 2 , the temperature parity ratio R T T , and antipodal correlation C ( π ) . Grey contours and one-dimensional distributions show the probability density obtained from Monte Carlo realizations of the observed sky. Red markers indicate the mean predictions of the SI model, while blue markers show the DSI prediction. The coloured dashed lines correspond to measurements from the four Planck component-separated maps (Commander, SMICA, SEVEM and NILC [26]), illustrating that the differences between foreground-cleaning methods are much smaller than the intrinsic cosmic variance. The observed realization lies much closer to the DSI prediction than to the SI expectation, giving a join Bayes factor B = 650 in favour of DSI.
Preprints 231015 g005
The large-scale CMB measurements indicate an asymmetry of order
A PT ≃ 0.2 .
This is also the characteristic magnitude of the low-multipole parity modulation predicted in the direct-sum construction. The effect decreases toward smaller angular scales, where the standard approximately parity-symmetric primordial spectrum is recovered.
Within this framework, A PT ≃ 0.2 should therefore be regarded as an observationally inferred parameter rather than merely as a qualitative characterization of a CMB anomaly. The evidence is not based only on the measured odd-parity excess itself. The same model simultaneously provides a better description of the low quadrupole, the even–odd power ratio and the antipodal correlation, with the joint analysis yielding
B DSI , SI ≃ 650 .
Thus the data favour the PT-asymmetric prediction over the parity-symmetric standard-inflation prediction by posterior odds of order several hundred to one, assuming comparable prior probabilities for the two models.
This distinction is important. The large cosmic variance of the lowest multipoles does not make A PT statistically meaningless. Rather, it is part of the predicted distribution against which the observation is compared. Within the primordial PT framework, the CMB therefore provides statistically significant evidence for a non-zero effective symmetry-breaking amplitude, with a preferred value of order
A PT ∼ 0.2 .
What remains model dependent is the physical interpretation of this measured asymmetry. Identifying A PT with symmetry breaking in the primordial quantum state assumes that the observed parity structure is cosmological and that the theoretical mapping between the underlying PT sectors and the CMB is correct. Foreground residuals, masks and estimator choices must therefore continue to be tested. However, these possible systematics should be distinguished from cosmic variance: the latter is already included in the likelihood calculation and does not account for the preference of one model over the other.
The next requirement is independent confirmation. If the inferred A PT is genuinely primordial, the same symmetry breaking should leave correlated signatures in other parity-sensitive observables, particularly CMB polarization and the primordial tensor sector. These provide genuine falsification tests of the interpretation rather than merely additional ways of increasing the statistical significance of the temperature result.

5.6. Beyond Temperature: Polarization and Tensor Modes

If the parity asymmetry is primordial, it should not be confined to the temperature map. The same underlying state should leave correlated signatures in CMB polarization and potentially in primordial gravitational waves.
Temperature and E-mode polarization are parity even, while B-mode polarization is parity odd. Cross-correlations such as T B and E B , which vanish in a parity-symmetric ensemble under standard assumptions, provide particularly sensitive tests of primordial parity violation.
A genuine PT-breaking mechanism should therefore predict a consistent pattern across
T T , T E , E E , T B , E B ,
rather than explaining only one statistic in the temperature sky.
The tensor sector provides an additional test. A PT-asymmetric primordial state may produce unequal amplitudes for opposite tensor helicities,
P h R ( k ) ≠ P h L ( k ) ,
leading to chiral primordial gravitational waves. Such a signal could appear in CMB polarization or, depending on the energy scale and transfer function, in future direct gravitational-wave observations.
The decisive test is therefore not the existence of one parity anomaly but cross-consistency among several independent parity-sensitive observables.

5.7. PT, CPT and the Origin of Matter

The possibility of primordial PT breaking has another potentially important consequence. Relativistic quantum field theory is expected to preserve CPT under broad conditions. If CPT remains exact at the fundamental level while the observed cosmological branch breaks PT, then the corresponding state must also distinguish charge conjugation.
Schematically,
C = ( CPT ) ( PT ) − 1 .
Thus a PT-asymmetric branch of an underlying CPT-symmetric theory may also be C asymmetric.
This observation does not by itself generate a baryon asymmetry. Charge-conjugation breaking is not sufficient if baryon and lepton numbers are exactly conserved. Actual baryogenesis requires processes that change B or L, together with the appropriate departure from equilibrium or equivalent dynamical conditions.
Nevertheless, primordial PT breaking can provide the symmetry bias that such processes require. During reheating, when the high-density gravitational state is converted into matter and radiation, the same asymmetry measured through the primordial perturbations may be transferred to particle populations. If baryon- or lepton-number-violating interactions are active, a PT-related asymmetry can then seed a non-zero B − L abundance, which electroweak sphalerons subsequently redistribute between baryons and leptons.
This possibility is especially intriguing because it connects two apparently unrelated observations: the parity structure of the CMB and the matter–antimatter asymmetry of the Universe.

5.8. An Observational Symmetry Parameter

The broader significance of A PT is therefore similar to that of ρ G discussed previously. Both quantities characterize microscopic properties of the high-density phase, but both may be inferred indirectly from late-time observations.
The emerging chain is
primordial quantum state ⟶ A PT ⟶ CMB parity , polarization , tensor chirality , matter - - antimatter asymmetry .
This turns discrete symmetry from a formal property of the quantum theory into an experimentally accessible parameter.
The next step is to connect this symmetry information with the dynamical scale of the high-curvature phase. Quadratic gravity provides a particularly useful example because the amplitude of scalar perturbations fixes the coupling α , while the same scalaron dynamics controls inflation and reheating. The combination of α and A PT therefore opens the possibility of linking the CMB amplitude, primordial parity and the observed baryon abundance within a single consistency framework.

6. Inflation and Reheating in Quadratic Gravity

Quadratic gravity provides a particularly economical realization of the inflationary phase discussed above. In a homogeneous and isotropic background the Weyl tensor vanishes, so the background dynamics of
S QG = ∫ d 4 x − g M Pl 2 2 R + α 2 R 2 + β 2 W μ ν ρ σ W μ ν ρ σ
is governed by the Einstein–Hilbert and R 2 terms. The resulting theory is dynamically equivalent to General Relativity coupled to an additional scalar degree of freedom, the scalaron.
This is important for a finite bouncing cosmology because the same high-curvature sector that becomes relevant near the bounce naturally generates an approximately de Sitter phase. On the expanding branch this appears as Starobinsky inflation, while the time-reversed contracting branch contains a corresponding deflationary phase. Inflation is therefore not inserted independently into the cosmological history: it emerges from the high-curvature gravitational dynamics.

6.1. R 2 Gravity and the Scalaron

The R + R 2 sector can be written in the conventional form
S = M Pl 2 2 ∫ d 4 x − g R + R 2 6 M 2 ,
where
α = M Pl 2 6 M 2 .
After a conformal transformation to the Einstein frame, the additional gravitational degree of freedom is represented by a canonical scalar field ϕ with potential
V ( ϕ ) = M Pl 4 8 α 1 − e − 2 / 3 ϕ / M Pl 2 .
At large ϕ the potential approaches a nearly constant plateau,
V 0 = M Pl 4 8 α ,
and the expansion becomes quasi-de Sitter. The scalaron therefore provides a purely gravitational mechanism for inflation without requiring the introduction of an additional fundamental inflaton field.
For a finite bouncing cosmology this same solution can be considered on both sides of the bounce. During contraction, the system approaches the high-curvature plateau through an approximately time-reversed quasi-de Sitter phase; after the turning point, it leaves the high-density state through inflationary expansion. To leading order the sequence is therefore
deflation ⟶ bounce ⟶ inflation .
The detailed matching through the bounce depends on the microscopic quantum-gravity state and cannot be inferred from the classical R 2 solution alone. Nevertheless, the existence of the quasi-de Sitter attractor on the expanding branch makes quadratic gravity a natural realization of the semiclassical dynamics introduced in Section 3.

6.2. The CMB Amplitude Fixes the Quadratic Coupling

A particularly attractive feature of R 2 inflation is its predictivity. The scalar perturbation amplitude is
A s ≃ V * 24 π 2 M Pl 4 ϵ * ,
where the star denotes evaluation when the pivot scale exits the Hubble radius.
For Starobinsky inflation, at N ≫ 1 e-folds before the end of inflation,
ϵ * ≃ 3 4 N 2 ,
which gives
A s ≃ N 2 144 π 2 α .
The observed value [27]:
A s ≃ 2.1 × 10 − 9
therefore determines
α ≃ N 2 144 π 2 A s ∼ 10 9 ,
for N ≃ 50 –60.
Equivalently, the scalaron mass is fixed to be
M = M Pl 6 α ≃ 1.3 × 10 − 5 M Pl ,
or approximately
M ∼ 3 × 10 13 GeV .
Thus the strength of the quadratic correction is not a free high-energy parameter once the primordial fluctuation amplitude is specified. A measurement made from CMB temperature and polarization fluctuations determines a coupling in the ultraviolet gravitational action.
The same model predicts, to leading order,
n s ≃ 1 − 2 N , r ≃ 12 N 2 ,
giving n s ≃ 0.964 and r ≃ 0.004 for N ≃ 55 . The observed scalar spectrum therefore constrains not only the inflationary phenomenology but the scale of the underlying high-curvature gravitational dynamics.

6.3. Inflation as the Expanding Side of the Bounce

In the conventional interpretation, Starobinsky inflation is treated as the earliest dynamically specified stage of the Universe. In the finite bouncing picture it instead describes the expanding side of a more complete history.
As discussed previously, the quasi-de Sitter behaviour generated by the R 2 sector has an approximately time-reversed counterpart during contraction. The inflationary expansion therefore restores, to first approximation, the evolution undergone during the preceding deflationary phase.
This distinction is particularly relevant for spatial curvature. The curvature constant k does not change during either phase, while its physical contribution scales as
k a 2 .
Deflation increases this contribution as the Universe contracts, and inflation subsequently suppresses it as the Universe expands.
Consequently, inflation need not be interpreted as the mechanism that determines why the global curvature of the finite Universe is small. If the contracting and expanding quasi-de Sitter phases are approximately symmetric, much of the flattening produced after the bounce simply reverses the curvature amplification produced before it. The residual curvature retains information about the original finite perturbation that underwent collapse.
This provides a conceptually different interpretation of near-flatness. A very large, weakly overdense perturbation can correspond to a small positive curvature from the outset. The present value of Ω k can then retain information about these initial conditions rather than being exponentially erased by an arbitrarily long period of one-sided inflation.

6.4. From Inflation to Reheating

Inflation ends when the scalaron leaves the flat part of the potential and begins to oscillate around its minimum. Near the minimum,
V ( ϕ ) ≃ 1 2 M 2 ϕ 2 ,
so the scalaron behaves approximately as a massive coherent field.
Its oscillations subsequently decay into ordinary quantum fields through gravitationally induced couplings. The energy stored in the high-curvature gravitational sector is thereby transferred to particles and radiation,
ρ gravity ⟶ ρ ϕ ⟶ ρ matter + ρ radiation .
Reheating therefore has a particularly transparent interpretation in quadratic gravity: it is the process through which the additional gravitational degree of freedom created by the R 2 term transfers its energy to the particle sector.
In the bouncing picture, this is also part of the broader quantum-to-classical transition. The high-density state first gives rise to an expanding quasi-de Sitter geometry and quantum fluctuations. Scalaron decay then populates the Universe with particles, while the primordial perturbations decohere and become the classical density perturbations from which later structure forms.
The reheating stage is therefore potentially important for more than the thermal history. It is also the epoch during which symmetry properties inherited from the primordial quantum state may be transferred to the matter sector.

6.5. A Possible Connection Between α and the Bounce Density

The fact that α is fixed observationally raises a particularly interesting question. Can the same coupling also determine the characteristic density ρ G of the high-density state?
The plateau of the scalaron potential defines a natural energy density,
ρ R 2 ∼ V 0 = M Pl 4 8 α .
For α ∼ 10 9 , this corresponds to an energy scale
ρ R 2 1 / 4 ∼ 10 16 GeV ,
of the familiar inflationary order.
The associated Hubble scale is
H R 2 2 ≃ V 0 3 M Pl 2 = M Pl 2 24 α .
This suggests the possible identification
ρ G ∼ ? M Pl 4 8 α .
Such a relation would be remarkable because it would connect the bounce density directly to the measured amplitude of primordial fluctuations:
A s ⟶ α ⟶ ρ G .
Using Equation (79),
1 α = 144 π 2 A s N 2 ,
so that
ρ G ∼ ? 18 π 2 A s N 2 M Pl 4 .
The characteristic bounce horizon would then also cease to be an arbitrary parameter:
R G = c H G = 3 c 2 8 π G ρ G .
It would ultimately be related to the CMB normalization through α .
This identification is not yet established. The scalaron plateau energy is the energy density of a high-field quasi-de Sitter configuration, whereas ρ G was introduced as the density of the semiclassical quantum-supported state reached during gravitational collapse. In particular, the minimum of the scalaron potential has zero potential energy; V 0 should therefore not be interpreted as the ground-state energy density of the scalaron itself.
The question is instead whether the same high-curvature dynamics that produces the R 2 plateau can determine the effective limiting density of the collapsing many-body system. Establishing this requires a description in which the quantum state of matter and the higher-curvature gravitational dynamics are treated consistently.
For the present purpose, Equation (91) should therefore be regarded as a falsifiable hypothesis rather than an assumption.

6.6. The Bounce Scale as an Observational Consistency Test

If ρ G can be predicted from α , an important observational loop closes. The CMB determines A s , which determines α . The quadratic coupling then predicts a characteristic high-density and causal scale,
A s ⟶ α ⟶ ρ G ⟶ R G .
But R G independently determines which structures can remain outside the causal horizon through the bounce. It therefore affects the characteristic masses and abundance of relic compact objects.
The same quantity could then be inferred in two very different ways:
A s ⟶ α ↓ ρ G ↑ relic abundance , SMBH seeds , GW events ⟶ bounce phenomenology .
Agreement between these independent determinations would provide a non-trivial test of the microscopic theory. Disagreement would falsify the proposed identification even if quadratic inflation itself remained successful.
This illustrates the broader strategy of treating the finite Universe as a laboratory for quantum gravity. A high-energy coupling is first measured indirectly through primordial fluctuations and then confronted with an entirely different set of observables associated with the bounce.

6.7. Reheating as a Bridge to the Matter Asymmetry

There is a second possible consistency relation hidden in the same dynamics. The measured scalar amplitude determines
1 α = 144 π 2 A s N 2 ,
while the CMB parity signal discussed in the previous section provides an independent estimate of
A PT ∼ 0.2 .
Reheating is the epoch in which the scalaron transfers gravitational energy to the particle sector. If the primordial PT asymmetry is also transferred during this process, then the combination
A PT α
provides a natural dimensionless quantity with which to characterize the resulting matter–antimatter asymmetry.
Numerically,
A PT α ∼ 0.2 10 9 ∼ 2 × 10 − 10 ,
remarkably close to the scale of the observed cosmological baryon asymmetry.
At this stage this correspondence is only suggestive. A quantitative relation requires a microscopic mechanism that transfers the PT asymmetry into a non-zero baryon or lepton number and must satisfy the relevant Sakharov conditions. In particular, PT or charge-conjugation breaking alone cannot generate a net baryon abundance if B and L remain exactly conserved.
Nevertheless, the numerical coincidence motivates the question developed in the next section: whether reheating in a PT-asymmetric high-curvature state can connect the independently measured quantities A s and A PT to the observed baryon abundance.

7. Origin of Matter: Baryogenesis and Leptogenesis

The observed Universe contains an overwhelming excess of matter over antimatter. This asymmetry is usually quantified either by the baryon-to-photon ratio,
η B ≡ n B − n B ¯ n γ ≃ 6 × 10 − 10 ,
or by the baryon yield,
Y B ≡ n B − n B ¯ s ≃ 8.6 × 10 − 11 ,
where s is the entropy density. Its value is accurately known from the abundance matching of light elements created from primordial Nucleosynthesis (see Figure 6).
The origin of this small but non-zero asymmetry remains one of the major open problems in cosmology and particle physics. In the finite bouncing framework developed here, the question takes on an additional dimension. If the primordial quantum state is PT symmetric but the observed expanding branch carries a measurable PT asymmetry, then the same symmetry breaking that appears in the CMB may also bias the production of particles and antiparticles during the transition from the high-curvature gravitational state to the hot radiation era.
This possibility does not remove the standard requirements for baryogenesis. Rather, it suggests that the primordial PT asymmetry may provide one of the ingredients needed to satisfy them.

7.1. The Sakharov Conditions

Sakharov identified three general conditions required for a dynamical generation of baryon asymmetry from initially symmetric conditions [28]:
baryon - number violation , C and CP violation , departure from thermal equilibrium .
The first condition is necessary because a net baryon number cannot be generated if baryon number is exactly conserved. The second distinguishes matter from antimatter, while the third prevents detailed balance from cancelling the generated asymmetry.
These conditions are remarkably general and underlie most mechanisms proposed for baryogenesis, including grand-unified baryogenesis, electroweak baryogenesis and leptogenesis [29].
The Standard Model itself contains baryon-number violation through non-perturbative electroweak processes. At sufficiently high temperature, transitions between gauge vacua change baryon and lepton number according to
Δ B = Δ L ,
and therefore violate
B + L ,
while conserving
B − L .
These processes, conventionally described through electroweak sphalerons, are efficient above the electroweak scale [30].
Thus the early Universe already contains a mechanism capable of changing baryon and lepton number. The remaining question is what produces the initial symmetry bias and, crucially, whether it generates a non-zero B − L abundance that can survive sphaleron processing.

7.2. PT Breaking, CPT Symmetry and Charge Asymmetry

The PT interpretation developed in the previous section suggests a possible origin for this bias.
Suppose that the fundamental theory preserves CPT, as expected for a local Lorentz-invariant quantum field theory, while the cosmological state emerging from the bounce spontaneously or dynamically selects one PT orientation. Then a PT-asymmetric state cannot simultaneously remain invariant under charge conjugation.
Schematically,
C = ( CPT ) ( PT ) − 1 .
If CPT remains exact while PT is broken by the selected semiclassical branch, the state will in general also distinguish particles from antiparticles.
This provides a potentially important link between two otherwise separate cosmological observations. The large-angle CMB parity asymmetry may measure the degree to which the primordial state departs from PT symmetry, while the baryon asymmetry measures the resulting imbalance in the matter sector.
However, the logical distinction is essential:
PT / C breaking ⇏ baryogenesis
if baryon and lepton numbers are exactly conserved.
PT breaking can provide a symmetry bias, but some interaction must still convert that bias into a non-zero B or L. Electroweak sphalerons, heavy-particle decays, Majorana neutrino interactions or other baryon- and lepton-number-violating processes can provide this second ingredient.
In this sense, the PT asymmetry does not replace the Sakharov mechanism. It may instead supply a cosmological origin for the C and CP asymmetry required by it.

7.3. Reheating as the Transfer from Gravity to Matter

Quadratic gravity provides a particularly natural epoch at which such a transfer could occur. At the end of R 2 inflation, the scalaron oscillates around the minimum of its potential and decays into matter and radiation. The energy stored in the gravitational sector is thereby converted into the particle content of the hot Big Bang. The schematic evolution is
high - curvature quantum state ⟶ scalaron oscillations ⟶ ⟶ particle production ⟶ radiation era .
If the primordial gravitational state carries a PT asymmetry A PT , reheating provides a natural stage at which this asymmetry could be transferred to the particle sector. The resulting matter–antimatter imbalance need not be generated instantaneously at the bounce itself. It may arise during scalaron decay or subsequently through interactions of the produced particles.
This observation is especially relevant because reheating is intrinsically a non-equilibrium process, thereby providing one of the Sakharov ingredients automatically.
The precise mechanism remains model dependent. Pure R + R 2 gravity coupled minimally to the Standard Model does not by itself guarantee the generation of baryon number. Additional baryon- or lepton-number-violating physics is still required. The point is instead that the finite-bounce scenario provides a natural symmetry bias and a natural non-equilibrium epoch in which this physics can operate.

7.4. Leptogenesis and the Role of B − L

One of the best-motivated possibilities is leptogenesis. In the original proposal of Fukugita and Yanagida, the out-of-equilibrium decays of heavy Majorana neutrinos generate a lepton asymmetry, which is subsequently converted partially into a baryon asymmetry by electroweak processes [31]. The essential quantity is not B + L but
B − L .
Electroweak sphalerons violate B + L while preserving B − L . Consequently, an asymmetry produced before the electroweak epoch survives most naturally if it contains a non-zero B − L component.
For the Standard Model particle content, when the relevant interactions and sphalerons are in equilibrium, the final baryon yield is related to the primordial B − L yield by [32]
Y B = 28 79 Y B − L .
This relation is central to the present discussion. It means that the problem can be separated into two stages:
primordial dynamics ⟶ Y B − L ,
followed by
Y B − L ⟶ Y B
through standard electroweak processing.
The finite-universe framework therefore does not need to generate the observed baryon abundance directly. It is sufficient for the quantum-to-classical transition to generate an appropriate B − L asymmetry. Standard electroweak physics can then convert part of this into baryon number. This is precisely the logic underlying conventional leptogenesis [31,33], but here the origin of the initial asymmetry may be tied to the PT structure of the primordial gravitational state.

7.5. A Possible Scaling with A PT and α

The results of the previous sections suggest an intriguing dimensional relation. The amplitude of primordial scalar fluctuations fixes the quadratic coupling,
1 α = 144 π 2 A s N 2 ,
while the large-angle CMB parity signal provides an independent estimate A PT ∼ 0.2 .
The relevant combination of A PT and α is not arbitrary. In quadratic inflation, α appears in the denominator of the primordial fluctuation amplitude,
A s ≃ N 2 144 π 2 α ,
and also sets the characteristic high-curvature scale,
M 2 = M Pl 2 6 α , V 0 = M Pl 4 8 α .
Thus 1 / α , rather than α , measures the effective strength of the primordial high-curvature dynamics. If A PT quantifies the fractional symmetry breaking of the primordial quantum state, a natural leading-order expectation is that the generated matter–antimatter asymmetry scales as the product of this fractional bias and the strength of the high-curvature process that transfers it to the matter sector:
Y B − L ∝ A PT 1 α .
The alternative product A PT α would have the opposite dependence: it would predict a larger asymmetry as the characteristic inflationary curvature scale becomes smaller. For a mechanism in which the asymmetry is generated during the same high-curvature transition that controls primordial perturbations and reheating, the inverse dependence on α is therefore the physically motivated one. The combination
A PT α
is therefore a small dimensionless quantity determined entirely by cosmological observables. Numerically,
A PT α ∼ 2 × 10 − 10 ,
for α ∼ 10 9 . This is strikingly close to the observed scale of the cosmological matter asymmetry. Motivated by this coincidence, consider the phenomenological ansatz
Y B − L = ξ A PT α ,
where ξ parametrizes the efficiency with which the primordial PT asymmetry is converted into a net B − L abundance during reheating or subsequent particle interactions. Using Equation (112) then gives
Y B = 28 79 ξ A PT α .
For
A PT ≃ 0.2 , α ≃ 10 9 ,
this gives
Y B ≃ 7.1 × 10 − 11 ξ .
Comparison with the observed value
Y B obs ≃ 8.6 × 10 − 11
corresponds to
ξ ∼ 1.2 .
Thus no extremely small or large efficiency factor is required at the level of this phenomenological estimate. Using the CMB normalization directly, Equation (122) can also be written as
Y B = 28 79 ξ A PT 144 π 2 A s N 2 .
This equation expresses a potentially remarkable consistency relation between three independently measured properties of the Universe:
A s , A PT , Y B .
The scalar perturbation amplitude determines the strength of the R 2 interaction, the parity structure of the CMB measures the primordial symmetry breaking, and the baryon abundance measures the resulting matter–antimatter asymmetry.

7.6. What Remains to Be Derived

Equation (121) is not yet a prediction of quadratic gravity or of PT-symmetric quantum theory. It is a phenomenological scaling motivated by the observed magnitudes of the relevant dimensionless quantities. A microscopic derivation must explain why the generated B − L yield is proportional to A PT , why the suppression scale is 1 / α , and what determines the efficiency parameter ξ .
This distinction becomes particularly important when comparing with conventional leptogenesis. If A PT / α represented only the CP asymmetry per decay of a heavy particle, the final abundance would generally contain additional suppression factors associated with the number density of the decaying particles, washout and thermal efficiency. In that interpretation, one would not expect ξ automatically to be of order unity.
For the direct scaling in Equation (121) to hold, A PT / α must instead characterize something closer to the comoving B − L yield produced by the gravitational-to-matter transition, or the reheating dynamics must compensate the usual abundance and efficiency factors.
The result should therefore be viewed as a target for the microscopic theory rather than as an established derivation.

7.7. A Cosmological Consistency Test

Despite this caveat, the relation is potentially powerful because its ingredients are independently measurable. The proposed chain is
A s ⟶ α ,
CMB parity ⟶ A PT ,
and
α , A PT + B / L violation ⟶ Y B − L ⟶ Y B .
The first two quantities are inferred from primordial perturbations, while the third is measured independently through primordial nucleosynthesis and the CMB baryon density.
If a microscopic theory can derive Equation (121) with an order-unity and calculable coefficient, the observed baryon abundance would become a quantitative test of the same high-curvature physics responsible for inflation and primordial PT breaking.
Conversely, if the required conversion efficiency is inconsistent with reheating, sphaleron dynamics or particle-physics constraints, the proposed connection would be falsified. The numerical proximity
Y B ∼ A PT α
should therefore not be regarded merely as numerology. It identifies a concrete theoretical problem: determine whether the quantum-to-classical transition of a PT-asymmetric quadratic gravitational state naturally generates a B − L yield of this order. The answer can ultimately be tested against three independent cosmological observables.

8. Relics of the Contracting Phase

A finite bouncing cosmology allows structure to have a history that predates the expanding hot Big Bang [7]. During the contracting phase, gravitational instability amplifies density perturbations and can drive sufficiently overdense regions into the nonlinear regime before the background reaches the high-density state. Some of these structures can subsequently become larger than the shrinking causal horizon, survive the bounce as super-horizon perturbations and collapse into black holes after horizon re-entry.
These objects provide a second source of compact seeds, physically distinct from primordial black holes generated from the inflationary fluctuation spectrum. We refer to them as relic black holes or bounce relics, because their nonlinear structure was already established during the contracting phase.
This distinction is important observationally. Inflation can continue to produce the nearly Gaussian, nearly scale-invariant perturbations measured in the CMB, while rare strongly nonlinear structures originate from an earlier epoch. A significant compact-object population therefore need not be generated from the extreme tail of the same perturbation distribution that produces the CMB.

8.1. Structure Formation During Contraction

The linear evolution of matter perturbations already illustrates why structure formation can become particularly efficient during contraction. In a matter-dominated background the density contrast contains two independent solutions [34,35],
δ L ( a ) = A D + ( a ) + B D − ( a ) ,
with
D + ( a ) ∝ a , D − ( a ) ∝ a − 3 / 2 .
During expansion, D + is conventionally called the growing mode, whereas D − rapidly decays. During contraction, however, the scale factor decreases with time. The mode D − ∝ a − 3 / 2 therefore grows rapidly toward the bounce and can eventually dominate the density field.
This behaviour can be incorporated into the standard statistical description of halo formation given by the spherical collapse (see Figure 7). The same modeling can be applied for the whole observable universe as a FLRM cloud within a parent background or to perturbations within that universe. We can define the variance on scale R given a matter power spectrum P ( k ) ,
σ 2 ( R , a ) = D 2 ( a ) 2 π 2 ∫ d k k 2 P ( k ) W 2 ( k R ) ,
where W is an appropriate smoothing window. Replacing the standard expanding growth factor by the contracting solution allows the usual Press–Schechter, Sheth–Tormen or related halo-mass-function methods to estimate the abundance of collapsed structures before the bounce,
d n h d M = F σ ( M , a ) , δ c .
Figure 7. Configuration of a spherical collapse perturbation. We consider three spherically symmetric regions: (1) an outer background with radius > r and mean density ρ ¯ ; (2) an inner region of radius R < r with higher density ρ = ρ ¯ ( 1 + δ ) ; and (3) an intermediate near-vacuum layer separating them. This can represent a collapsing overdensity of size R (a halo) within a larger universe of size r, or analogously, our finite Universe of radius R embedded in a larger space.
Figure 7. Configuration of a spherical collapse perturbation. We consider three spherically symmetric regions: (1) an outer background with radius > r and mean density ρ ¯ ; (2) an inner region of radius R < r with higher density ρ = ρ ¯ ( 1 + δ ) ; and (3) an intermediate near-vacuum layer separating them. This can represent a collapsing overdensity of size R (a halo) within a larger universe of size r, or analogously, our finite Universe of radius R embedded in a larger space.
Preprints 231015 g007
The rapidly growing contraction mode enhances nonlinear structure formation as the bounce is approached, particularly at the high-mass end of the halo population. A first implementation of this calculation confirms that contracting cosmologies can therefore contain a substantial population of massive halos before entering the high-density phase.

8.2. Virialization and Nonlinear Relic Seeds

Once a perturbation becomes nonlinear, its evolution progressively decouples from that of the homogeneous background. In the spherical-collapse description, gravitationally bound dark-matter halos virialize at a characteristic overdensity of order
Δ vir = 1 + δ vir ∼ 200 ,
with the precise value depending on cosmology and on the convention used to define the virial radius [36].
As a measure of the degree of nonlinearity, this is far larger than the characteristic density contrast associated with PBH collapse at horizon entry,
δ PBH ∼ 0.3 − 0.7 ,
where the precise value depends on the equation of state, the definition of the perturbation amplitude and the shape of the perturbation profile [37,38].
This comparison does not imply that every virialized halo immediately becomes a black hole. A halo is a gravitationally bound many-body system, not necessarily a compact object. The relevant point is instead that a halo formed during contraction can already possess a strongly nonlinear density contrast before it becomes super-horizon. If this nonlinear configuration survives the bounce and later re-enters the causal horizon, it can satisfy the conditions for rapid gravitational collapse without having originated from an exceptionally rare inflationary fluctuation.
The contracting phase can therefore generate nonlinear compact-object seeds while leaving the inflationary perturbation spectrum itself close to Gaussian.

8.3. Horizon Exit and Survival Through the Bounce

The causal scale during contraction is characterized by
R H = c | H | .
As the high-density state is approached, | H | grows and the Hubble radius shrinks. A sufficiently large halo can therefore reach
R halo > R H
and become super-horizon.
The smallest causal scale reached close to the bounce is controlled by the ground-state density ρ G . In the semiclassical description introduced above,
R B ≃ c H G = 3 c 2 8 π G ρ G ,
where
H G 2 = 8 π G 3 ρ G .
The value of ρ G therefore acts as a causal filter on the pre-bounce structure population. Perturbations whose physical scales become larger than the shrinking horizon can survive as effectively frozen super-horizon configurations, whereas the evolution of structures that remain inside the horizon throughout the high-density phase depends directly on the microscopic quantum dynamics and cannot be determined reliably within the present semiclassical treatment.
Using the characteristic stellar degeneracy density as a conservative lower limit,
ρ G ≳ ρ SD ,
gives approximately
R B ≲ 24 km ,
corresponding to a gravitational mass scale of order
M B ≲ 8.3 M ⊙ .
These values should be regarded only as conservative reference scales. If the actual ground-state density is substantially larger, both the minimum causal scale and the corresponding mass scale become much smaller.

8.4. Horizon Re-Entry and Relic Black Holes

After the bounce, the causal horizon grows again. Pre-bounce structures that survived as super-horizon perturbations eventually re-enter causal contact.
Their initial conditions at re-entry differ fundamentally from those of ordinary inflationary perturbations. A standard primordial fluctuation typically re-enters while still in or close to the linear regime and must subsequently exceed the appropriate collapse threshold. A bounce relic, by contrast, may already have undergone nonlinear collapse and virialization before leaving the horizon.
The two formation channels can therefore be summarized as
standard PBH : inflationary fluctuation → horizon re - entry → nonlinear collapse , bounce relic : pre - bounce halo → nonlinear evolution → horizon exit → bounce → horizon re - entry → black - hole collapse .
The relic channel therefore separates the production of compact objects from the statistics of inflationary fluctuations. The post-bounce matter field can contain two physically different components,
δ ( x ) = δ inf ( x ) + δ relic ( x ) ,
where δ inf is approximately Gaussian and nearly scale invariant, whereas δ relic represents a sparse population of strongly nonlinear structures inherited from contraction.

8.5. From the Pre-Bounce Halo Population to the Relic Abundance

The abundance of relic black holes need not be introduced as an arbitrary parameter. Once a statistical description of the perturbations during contraction is specified, the relic mass function can be estimated from the pre-bounce halo population.
A particularly simple exploratory model illustrates this connection. Suppose that the contracting matter spectrum can be approximated as
P ( k ) = A k n .
For a fixed spectral index n, the amplitude A can be parametrized by a variance σ 8 , defined in analogy with the usual late-time normalization on 8 h − 1 Mpc scales. Here σ 8 should be understood as a parameter describing the amplitude of the pre-bounce fluctuation spectrum, not necessarily as the present measured value.
The curvature scale χ k provides another independent parameter. In the finite-universe construction it characterizes the positive curvature, or equivalently the binding energy, of the original collapsing perturbation. It therefore affects the background evolution during contraction.
The third relevant quantity is the ground-state density ρ G . This determines how close to the bounce the semiclassical phase extends and, through the bounce dynamics, the epoch at which increasingly large structures leave the shrinking causal horizon.
The halo mass function can therefore be evaluated at a characteristic freeze-out epoch a G ,
d n relic d M ≃ d n h d M a = a G ,
where
a G = a G ( ρ G , χ k ) ,
subject to the additional requirement that the relevant structures become super-horizon and survive the bounce.
The total relic density follows from
Ω relic = 1 ρ crit ∫ d M M d n relic d M .
The general dependence can therefore be written schematically as
Ω relic = Ω relic ρ G , σ 8 , n , χ k .
This is an important result because it transforms the relic population from a qualitative possibility into a calculable consequence of the contracting cosmology. If the spectral shape and curvature are fixed, the relation reduces to
Ω relic = Ω relic ( ρ G , σ 8 )
within the assumptions of the model.
A first calculation using a power-law spectrum and representative curvature scales finds a strong, approximately power-law degeneracy between these two parameters along contours of fixed relic abundance [39],
log ρ G ρ SD ≃ A log σ 8 + B ,
with a slope close to A ≃ − 4 3 over the range explored.
The normalization depends on the assumed spectral index, curvature scale and other simplifying assumptions, so this numerical relation should not yet be regarded as a fundamental prediction. Its importance is conceptual: the observed relic abundance constrains a combination of the microscopic bounce density and the amplitude of structure present during contraction.

8.6. Relics as Dark Matter

If the relic black holes survive to the present epoch, their density must satisfy
Ω relic ≤ Ω m ,
and, if they are interpreted as a dark-matter population, more specifically
Ω relic ≤ Ω c .
The strongest version of the scenario corresponds to
Ω relic ≃ Ω c ,
so that relic black holes account for most or all of the cold dark matter. However, the BHU framework does not require this equality; relics may constitute only a fraction of the dark matter. The important point is that the condition
Ω relic ≤ Ω c
already removes part of the ( ρ G , σ 8 ) parameter space. In the exploratory calculation described above, regions that predict too many relics are directly excluded. The observed dark-matter abundance therefore provides a quantitative constraint on the ground-state density even before other compact-object observations are included.
This conclusion is stronger than simply noting that relic black holes could be dark matter. Their predicted abundance is connected to the physical parameters controlling the contraction and bounce.
Additional constraints arise from the relic mass distribution. Depending on their masses and abundance, compact relics can be probed through gravitational lensing, accretion, dynamical effects, binary merger rates, CMB energy injection and stochastic gravitational-wave backgrounds. A full comparison with these constraints requires the relic mass function rather than only its integrated density.

8.7. Seeds of Supermassive Black Holes

The high-mass tail of the relic population provides a second and largely independent observational test. Quasars powered by black holes of order 10 9 M ⊙ are already observed at z ≳ 7 , when the Universe was less than a billion years old [40]. Their existence places strong demands on conventional seed-formation scenarios, which must form sufficiently massive progenitors at very early times and sustain rapid growth through accretion and mergers [40,41].
In the bounce-relic picture, some of the nonlinear evolution required to form these seeds can occur before the hot Big Bang. Massive pre-bounce halos can leave the horizon, survive the bounce and later re-enter as strongly nonlinear perturbations. Their subsequent collapse provides an early population of massive black-hole seeds.
The relevant observable is therefore not only the total relic density but the cumulative number density above a specified mass,
n relic ( > M seed ) = ∫ M seed ∞ d M d n relic d M .
Figure 8 illustrates the a first exploration by [39]. This estimate is degenerate in the ( ρ G , σ 8 ) plane. The values shown limits the density but not the volume of the BHU. The figure shows that relics with initial masses
M relic ≳ 10 4 M ⊙
can, within regions of parameter space compatible with the total dark-matter abundance, be sufficiently numerous to provide seeds for the observed supermassive-black-hole population, provided subsequent growth through accretion and mergers is efficient.
The abundance of very massive relics therefore probes a different part of the relic mass function from the total dark-matter density. This distinction is important because the two measurements can begin to break the degeneracy between ρ G and the initial fluctuation amplitude.

8.8. Relics as Probes of the Bounce and the Contracting Phase

The relic population ultimately contains information about more than the bounce itself. Its abundance and mass distribution result from the combination of four conceptually distinct ingredients:
σ 8 , n , χ k , ρ G ⟶ d n relic d M .
The first two quantities characterize the amplitude and scale dependence of density perturbations during contraction. The curvature scale χ k characterizes the binding energy and global initial conditions of the finite perturbation. The density ρ G characterizes the microscopic high-density state responsible for the bounce.
Different observations probe different integrals or regions of the same mass function,
d n relic d M ⟶ Ω relic , n SMBH , lensing , binary mergers , stochastic GW backgrounds .
With sufficiently many independent observables, the degeneracies among the parameters can therefore be broken. The total relic density constrains the integrated mass function, the abundance of early massive black holes probes its high-mass tail, while lensing and gravitational waves probe different mass intervals.
This opens an inverse problem of particular interest:
present relic observations ⟶ ρ G , σ 8 , n , χ k pre − bounce .
Late-time observations could therefore constrain simultaneously the microscopic endpoint of the collapse and the classical conditions from which the collapse began.
The curvature scale is especially interesting in this respect. As discussed earlier, positive curvature measures the gravitational binding of the original finite perturbation. Constraining χ k together with the pre-bounce power spectrum would therefore provide information about the size and initial overdensity of the perturbation from which our finite Universe formed and, potentially, about the parent background in which that perturbation originated.
Relic black holes would then be fossils not only of the quantum bounce but also of the pre-bounce Universe.

8.9. A Cross-Consistency Test of the Quantum-Gravity Scale

The relic calculation becomes even more powerful if the ground-state density can be predicted independently from the microscopic theory.
In the quadratic-gravity example discussed above, we proposed the possibility
ρ G ∼ ? M Pl 4 8 α .
Since the CMB scalar amplitude determines α ,
A s ⟶ α ,
the high-curvature theory would then predict ρ G independently of the relic calculation.
The same parameter could consequently be constrained along two entirely different observational routes:
A s ⟶ α ⟶ ρ G ,
and
Ω relic , n SMBH , lensing , GW observations ⟶ ρ G
after marginalizing over the properties of the pre-bounce perturbation spectrum.
Agreement would connect the amplitude of primordial quantum fluctuations, the microscopic high-curvature gravitational action and the late-time population of compact objects. Disagreement would falsify the proposed identification of the quadratic-gravity scale with the density responsible for the bounce.
The relic population is therefore not merely an additional phenomenological consequence of a bouncing cosmology. Under relatively simple assumptions, its abundance is calculable and provides a quantitative connection between present observations, the conditions during gravitational collapse and the microscopic physics of the high-density state.

9. Strong-Gravity and Gravitational-Wave Tests

The cosmological observables discussed so far probe quantum gravity indirectly through the primordial Universe and the relic population inherited from the contracting phase. Black holes provide a complementary laboratory. Their near-horizon geometry accesses strong gravitational fields in the late Universe and can therefore test aspects of the same microscopic theory that controls the high-curvature dynamics of the bounce.
Figure 9. Quasi-normal ringing observed in the gravitational-wave signal from the first detected binary black-hole merger, GW150914. After the merger the newly formed black hole settles to equilibrium through damped oscillations known as quasi-normal modes. The frequency and decay time of the oscillation depend only on the mass and spin of the final black hole and provide a direct test of general relativity in the strong-field regime. Adapted from [43].
Figure 9. Quasi-normal ringing observed in the gravitational-wave signal from the first detected binary black-hole merger, GW150914. After the merger the newly formed black hole settles to equilibrium through damped oscillations known as quasi-normal modes. The frequency and decay time of the oscillation depend only on the mass and spin of the final black hole and provide a direct test of general relativity in the strong-field regime. Adapted from [43].
Preprints 231015 g009
This complementarity is particularly important for quadratic gravity. In an exactly homogeneous FLRW background the Weyl tensor vanishes, so the background cosmological dynamics are controlled primarily by the R 2 sector. Around black holes, by contrast, the Weyl curvature is non-zero and the spin–2 sector becomes directly relevant [21,22]. Cosmology and black-hole observations therefore probe different components of the same gravitational action.
The basic observational strategy is to look for departures from the predictions of General Relativity in systems where the gravitational field is strongest. Particularly promising probes include black-hole ringdown, quasinormal-mode spectroscopy, horizon absorption, extreme-mass-ratio inspirals and possible parity-dependent signatures in gravitational waves [44,45,46].

9.1. Black-Hole Quasinormal Modes

After a perturbed black hole settles toward equilibrium, the emitted gravitational radiation is dominated by a discrete set of damped oscillations known as quasinormal modes [44]. Their complex frequencies can be written as
ω ℓ m n = ω ℓ m n R − i ω ℓ m n I ,
where ω R determines the oscillation frequency and ω I the decay rate.
In General Relativity, the quasinormal spectrum of an isolated Kerr black hole is determined entirely by its mass and angular momentum. Measuring several independent modes therefore provides a direct consistency test of the black-hole geometry and forms the basis of black-hole spectroscopy [45].
Higher-curvature modifications can alter both the background solution and the equations governing perturbations. In quadratic gravity, the Weyl-squared sector introduces additional spin–2 dynamics and can modify the quasinormal spectrum even when the large-scale geometry remains close to that of General Relativity [21,22].
The observational question is therefore whether the measured frequencies and damping times satisfy
ω ℓ m n obs = ω ℓ m n GR ( M , J )
for a single pair ( M , J ) .
Any statistically significant inconsistency between multiple measured modes would indicate either additional degrees of freedom or a modification of the near-horizon gravitational dynamics.

9.2. Parity and Isospectrality

For a non-rotating Schwarzschild black hole, linear gravitational perturbations separate into axial and polar sectors. The axial sector is described by the Regge–Wheeler equation, while the polar sector is described by the Zerilli equation [47,48]. Although the two effective potentials are different, they possess the same quasinormal-mode spectrum,
ω ℓ n axial = ω ℓ n polar .
This remarkable isospectrality reflects a hidden relation between the two perturbation equations and is discussed in detail in the classic treatment of black-hole perturbations by Chandrasekhar [49]. It therefore provides a particularly clean probe of the symmetry structure of the gravitational theory.
Modified gravity can break this relation. Extra scalar or tensor degrees of freedom can couple differently to the two parity sectors, producing
ω ℓ n axial ≠ ω ℓ n polar .
The connection with the PT discussion developed earlier should be made with care. Schwarzschild isospectrality is not itself a direct consequence of PT symmetry. Nevertheless, it provides an independent strong-gravity test of whether parity-related gravitational sectors retain the symmetry structure predicted by General Relativity or acquire modifications associated with the underlying quantum theory.
Future measurements of multiple ringdown modes may therefore test not only the Kerr geometry but also the parity structure of gravitational perturbations.

9.3. Additional Gravitational Degrees of Freedom

The quadratic action
S QG = ∫ d 4 x − g M Pl 2 2 R + α 2 R 2 + β 2 W 2
contains, in addition to the usual massless graviton, a scalar mode associated with R 2 and a spin–2 sector associated with W 2 [21,22].
The scalar mode has characteristic mass
M 0 2 ≃ M Pl 2 6 α ,
while the characteristic scale of the additional spin–2 mode is set schematically by
M 2 2 ∼ M Pl 2 | β | ,
up to convention-dependent numerical factors.
If these modes can be excited in strong-gravity systems, they may produce additional resonances or modify the propagation and damping of the standard gravitational modes.
The precise prediction depends crucially on the quantization of the spin–2 sector. In the conventional treatment it is associated with the higher-derivative ghost problem. Alternative quantizations, including the direct-sum construction discussed above, modify the interpretation of this sector while retaining the ultraviolet improvement of the quadratic action.
Gravitational-wave spectroscopy therefore offers a direct observational test of whether such additional degrees of freedom are physically realized.

9.4. Horizon Absorption

Another clean strong-gravity observable is the absorption of gravitational radiation by a black-hole horizon.
In General Relativity, part of the gravitational-wave flux produced by an orbiting object is absorbed by the horizon [50]. The corresponding contribution to the orbital evolution can be written schematically as
E ˙ = E ˙ ∞ + E ˙ H ,
where E ˙ ∞ is the flux radiated to infinity and E ˙ H is the flux entering the horizon.
The magnitude and sign of E ˙ H depend on the black-hole spin and the frequency of the perturbation. For rapidly rotating black holes, superradiant interactions can even make the effective horizon flux negative [51].
Any modification of the microscopic horizon structure may alter this absorption. A perfectly reflecting or partially reflecting quantum boundary, for example, would give a different flux from a classical absorbing horizon [46].
Measurements of horizon absorption therefore provide a direct probe of whether astrophysical black holes possess the dissipative properties predicted by General Relativity.

9.5. Extreme-Mass-Ratio Inspirals

Extreme-mass-ratio inspirals provide one of the most precise future probes of black-hole geometry [52,53]. In these systems, a stellar-mass compact object orbits a massive black hole for thousands to millions of cycles before merger.
The accumulated gravitational-wave phase is extremely sensitive to the background geometry and to small corrections in the radiation-reaction force. Schematically,
Φ GW = ∫ Ω orb d t ,
so that even a very small fractional modification of the orbital evolution can accumulate into an observable phase shift over the full inspiral.
EMRIs can therefore test several aspects of strong gravity simultaneously: the multipole structure of the central object, horizon absorption, additional gravitational degrees of freedom and possible deviations from the Kerr metric.
This makes them particularly complementary to ringdown observations. Ringdown probes the response of the black hole after a violent perturbation, whereas an EMRI maps the geometry and dissipative properties of the same object over a long period of quasi-adiabatic evolution.
These systems are among the principal science targets of LISA [? ], which should provide access to long-duration inspirals into massive black holes and thereby substantially improve tests of near-horizon gravity.

9.6. Primordial and Astrophysical Gravitational Waves

Gravitational waves also connect the strong-gravity and cosmological components of the framework.
A high-curvature bounce and the subsequent inflationary phase can produce a primordial tensor background,
P h ( k ) ,
whose amplitude, scale dependence and parity structure depend on the microscopic gravitational theory.
If the primordial state breaks PT symmetry, the two tensor helicities may have different spectra,
P h R ( k ) ≠ P h L ( k ) .
Such chirality can be tested through CMB polarization and, depending on the frequency range and transfer function, potentially through direct gravitational-wave observations.
At later times, the relic population discussed in the previous section can generate an independent astrophysical gravitational-wave background through binary formation and mergers.
The total signal may therefore contain two conceptually different components,
Ω GW ( f ) = Ω GW primordial ( f ) + Ω GW relic ( f ) .
The first probes the quantum state and high-curvature dynamics directly, while the second probes the surviving population of structures formed during contraction.

9.7. A Hierarchy of Strong-Gravity Tests

The strongest conclusions will not come from any single departure from General Relativity but from consistency among several independent observables.
For black holes, the relevant chain is
microscopic gravitational theory ⟶ QNM spectrum , axial - - polar relations , horizon absorption , EMRI phasing , additional GW modes .
Each observable probes a different aspect of the same near-horizon physics. For example, a modification that shifts the quasinormal spectrum but leaves horizon absorption exactly unchanged would have to be consistent with the same underlying perturbation equations. Likewise, an additional spin–2 degree of freedom inferred from ringdown should also affect propagation or orbital dynamics in a predictable way.
This creates a network of internal consistency tests that is much harder to satisfy accidentally than a single anomalous measurement.

9.8. Connecting Black Holes and the Cosmological Bounce

The broader motivation for including strong-gravity tests in the present review is that the cosmological bounce and astrophysical black holes probe different regimes of the same gravitational theory.
In quadratic gravity, for example,
R 2
dominates aspects of the homogeneous high-curvature cosmology, while
W μ ν ρ σ W μ ν ρ σ
becomes especially important in anisotropic and strongly inhomogeneous regions such as black-hole spacetimes [21,22].
The observational programme is therefore naturally divided into two complementary channels,
cos mology ⟶ { α , ρ G , A PT , primordial spectra } , black holes ⟶ { β , spin - - 2 dynamics , horizon structure } .
A complete quantum-gravity theory must describe both consistently.
This is one of the principal advantages of treating the finite Universe as a laboratory rather than focusing on a single observable. The same microscopic theory can be tested in the primordial high-density state, through relic populations and around astrophysical horizons in the late Universe. A successful model must reproduce all three regimes with one consistent set of degrees of freedom and couplings.

10. Cross-Consistency Tests

The central claim of this review is not that any single observation provides a decisive signature of quantum gravity. The stronger possibility is that a finite bouncing cosmology allows several independent observations to constrain the same underlying physical quantities.
This distinction is essential. A sufficiently flexible model can often be adjusted to reproduce one anomalous observation. It is much harder for the same microscopic theory, with the same parameters, to account simultaneously for the primordial scalar spectrum, CMB parity, the baryon asymmetry, spatial curvature, the relic compact-object population and strong-gravity gravitational-wave observations.
The finite-Universe framework therefore naturally leads to a programme of cross-consistency tests. Rather than asking whether each observable is individually compatible with the model, one asks whether independent measurements infer mutually consistent values of the same physical parameters.
The principal quantities introduced in the previous sections are
α , β , A PT , ρ G , σ 8 pre , n pre , χ k ,
where α and β characterize the quadratic gravitational action, A PT characterizes the primordial symmetry breaking, ρ G defines the high-density state, and { σ 8 pre , n pre , χ k } characterize the conditions of the contracting finite perturbation.
These parameters are not all fundamental in the same sense. Some may ultimately be derived from a more complete microscopic theory, while others encode cosmological initial conditions. The important point is that they enter several independent observational channels.

10.1. The Inflationary Normalization: A s → α

The first relation is already well established within R 2 inflation. The amplitude of scalar fluctuations satisfies
A s ≃ N 2 144 π 2 α ,
so that
α ≃ N 2 144 π 2 A s .
The measured value
A s ≃ 2.1 × 10 − 9
therefore fixes
α ∼ 10 9
for the usual range of inflationary e-folds.
This is the first important step in the observational programme: a dimensionless coupling in the high-curvature gravitational action is determined by a late-time measurement of primordial fluctuations.
The same value predicts the scalaron mass,
M 0 = M Pl 6 α ,
and the characteristic inflationary energy scale,
V 0 = M Pl 4 8 α .
Thus several apparently different properties of the primordial Universe are already tied to a single measured parameter.

10.2. A First Consistency Loop: A s , α and ρ G

If the high-density state responsible for the bounce is controlled by the same quadratic dynamics, one may hypothesize
ρ G ∼ ? M Pl 4 8 α .
This relation is not required by the finite-bounce construction and has not yet been derived microscopically. It is therefore an ideal consistency test.
The CMB predicts
A s ⟶ α ⟶ ρ G QG ,
while the relic population provides an independent route,
Ω relic , n SMBH , lensing , GW mergers ⟶ ρ G relic ,
after marginalizing over the pre-bounce perturbation spectrum.
The comparison
ρ G QG = ? ρ G relic
is therefore a genuine cross-consistency test.
Agreement would imply that the scale inferred from CMB quantum fluctuations also determines the causal filtering of nonlinear structures through the bounce. Disagreement would rule out the proposed identification even if R 2 inflation remained a successful description of the primordial scalar spectrum.

10.3. A Second Consistency Loop: A s , A PT and Y B

A conceptually different test connects the amplitude of primordial fluctuations to the matter–antimatter asymmetry.
The CMB scalar amplitude determines
1 α = 144 π 2 A s N 2 ,
while large-angle parity measurements provide an independent estimate
A PT ≃ 0.2 .
If the PT asymmetry of the primordial state is transferred to a non-zero B − L abundance during reheating, the phenomenological relation introduced above is
Y B − L = ξ A PT α .
Standard electroweak sphaleron processing then gives
Y B = 28 79 ξ A PT α ,
or equivalently
Y B = 28 79 ξ A PT 144 π 2 A s N 2 .
The three quantities
A s , A PT , Y B
are measured through physically different phenomena. Their possible relation therefore cannot be reproduced by adjusting a single observable after the fact.
At present the unknown coefficient ξ prevents this from being a complete prediction. The crucial theoretical task is to derive ξ from reheating and baryon- or lepton-number-violating microphysics.
If this calculation yields a value of order unity, the measured baryon abundance would become an independent confirmation of the same primordial symmetry breaking inferred from the CMB. If it requires an implausibly large or small efficiency, the proposed connection would fail.

10.4. A Third Consistency Loop: Curvature and the Initial Collapse

Spatial curvature provides a qualitatively different constraint because it does not primarily probe the high-density quantum state. Instead, it retains information about the initial conditions of the finite perturbation that underwent gravitational collapse.
The curvature scale can be written as
k = 1 χ k 2 ,
with
Ω k = − k c 2 a 2 H 2 .
In the finite-bounce picture, post-bounce inflation does not generate the underlying value of k. The preceding deflationary phase enhances the physical curvature contribution, while the subsequent inflationary phase approximately reverses this evolution. The comoving curvature remains fixed.
The observed value of Ω k therefore probes the binding energy of the original collapsing perturbation.
This interpretation can be tested independently through relics because the same curvature scale enters the pre-bounce background evolution and hence the relic mass function,
Ω relic = Ω relic ρ G , σ 8 pre , n pre , χ k .
Thus the curvature inferred directly from cosmological geometry can be compared with the curvature required to reproduce the observed relic population:
χ k geometry = ? χ k relic .
This is particularly interesting because χ k is related to the initial binding energy of the perturbation. A consistent determination would therefore constrain not only the bounce but the conditions under which the finite Universe originally separated from its parent background.

10.5. The Relic Consistency Triangle

The relic population itself contains several independent observables.
The total density measures
Ω relic = 1 ρ crit ∫ d M M d n relic d M ,
while the abundance of early massive black holes probes
n relic ( > M seed ) = ∫ M seed ∞ d M d n relic d M .
Lensing and gravitational-wave observations probe other mass intervals and different moments of the same distribution. These observables form a natural consistency triangle,
GW / lensing / ∖ n SMBH − − Ω relic
all of which must be reproduced by a single relic mass function. This is more restrictive than just requiring
Ω relic ≤ Ω c .
A mass function can satisfy the total dark-matter abundance while overproducing massive seeds, merger events or lensing signatures. Conversely, a population capable of explaining early SMBHs may contribute only a small fraction of the total dark matter.
The complete relic model must therefore satisfy simultaneously
Ω relic , n SMBH , Γ merger , τ lens , Ω GW relic predicted = … observed .
This provides multiple constraints on the same pre-bounce parameters { ρ G , σ 8 pre , n pre , χ k } .

10.6. The PT Consistency Triangle

The PT interpretation likewise predicts that the parity asymmetry should not appear only in one CMB statistic.
If the observed temperature parity is primordial, the same underlying A PT should generate correlated effects in
T T , T E , E E , T B , E B ,
and potentially in the helicity structure of primordial tensor modes.
The corresponding consistency structure is
A PT T T / ∖ A PT pol − − A PT tensor .
A primordial explanation requires these independent measurements to point toward a common symmetry-breaking amplitude,
A PT T T ≃ A PT pol ≃ A PT tensor ,
up to the transfer functions predicted by the microscopic theory.
This provides a much stronger test than increasing the statistical significance of the temperature anomaly alone. Failure of the predicted polarization or tensor correlations would challenge the interpretation of the temperature parity signal as primordial PT breaking.

10.7. The Strong-Gravity Consistency Triangle

A similar structure arises in black-hole physics.
The Weyl sector of quadratic gravity can affect
QNM frequencies , isospectrality , horizon absorption , EMRI phasing .
These phenomena are measured in physically different regimes but depend on the same strong-field gravitational dynamics. Schematically,
horizon reflexion / absortion / ∖ EMRI − − − − − − − ringdown .
A modification of the spin–2 sector capable of shifting the ringdown spectrum must also yield a consistent prediction for orbital dynamics and horizon dissipation.
This allows gravitational-wave observations to constrain the same parameter β , or more generally the same microscopic modification of the spin–2 sector, through several independent measurements.

10.8. Connecting the Ricci and Weyl Sectors

Quadratic gravity makes the global consistency programme especially clear because its two curvature-squared terms dominate in different physical environments.
The homogeneous primordial Universe has
W μ ν ρ σ = 0 ,
so the background dynamics primarily constrain the Ricci sector and the coupling α .
Strongly inhomogeneous black-hole spacetimes have substantial Weyl curvature, making them sensitive to the coupling β .
The same microscopic theory must therefore satisfy
α ⟶ A s , n s , r , ρ G ? , Y B ? , β ⟶ QNM , EMRI , horizon absorption , tensor dynamics .
The cosmological and black-hole sectors are therefore complementary rather than redundant. Together they test the full quadratic action over a much larger range of geometries than either could probe alone.

10.9. An Observational Network

The complete framework can be summarized as a network shown in Figure 10 rather than a linear chain. The arrows in this diagram do not all represent established theoretical relations. Some are predictions of well-developed components of the framework, others are hypotheses that require microscopic derivation. That distinction is precisely what makes the network useful: each proposed connection can be tested independently.
A successful microscopic model must produce one mutually consistent point in this multidimensional observational space.

10.10. From Parameter Fitting to Over-Constraining the Theory

The objective is therefore not simply to fit the parameters
α , β , A PT , ρ G , σ 8 pre , n pre , χ k
independently.
The stronger goal is to reduce the number of independent parameters by deriving relations among them from the microscopic theory.
For example,
A s → α → ρ G
would remove ρ G as an independent bounce parameter, while a derivation of
Y B − L = ξ A PT α
with calculable ξ would connect the matter asymmetry to two quantities already measured in the CMB.
Similarly, a theory of the parent perturbation could predict
σ 8 pre , n pre , χ k ,
turning the relic mass function from a parametrized model into a genuine prediction.
The more such relations can be derived, the more the framework becomes over-constrained.
This provides a practical definition of falsifiability in the present context. The theory does not succeed because it can accommodate each observation separately. It succeeds only if the same microscopic dynamics predicts all of them simultaneously.

10.11. What Would Falsify the Framework?

The cross-consistency approach also makes clear how the framework can fail.
Examples include:
  • a value of ρ G inferred from relic observations that is incompatible with the scale predicted by quadratic gravity;
  • a primordial PT amplitude inferred from temperature anisotropies that is not reproduced in polarization or tensor observables;
  • a microscopic reheating calculation that predicts a baryon asymmetry incompatible with the observed Y B ;
  • a relic population capable of accounting for dark matter but inconsistent with lensing, merger rates or the observed SMBH abundance;
  • a measured spatial curvature incompatible with the curvature required by the pre-bounce relic population;
  • strong-gravity observations that exclude the spin–2 dynamics required by the quantum completion of quadratic gravity.
No single one of these tests is guaranteed to be decisive in isolation. Their power comes from probing different physical epochs and length scales.

10.12. The Universe as a Laboratory

The resulting observational programme spans an extraordinary hierarchy of scales. The scalar amplitude and CMB parity probe primordial quantum fluctuations. The baryon abundance probes the conversion of the high-curvature state into matter. Spatial curvature and the relic mass function probe the global collapse that preceded the bounce. Early supermassive black holes, lensing and binary mergers probe the surviving nonlinear structures. Ringdown and EMRIs probe the strong-field dynamics of astrophysical horizons.
These observations are separated by many orders of magnitude in length, energy and cosmic time, yet in a finite bouncing cosmology they need not be independent. They can instead be different experimental windows onto the same underlying gravitational theory.
This is the sense in which a finite Universe can act as a laboratory for quantum gravity: not because quantum-gravitational effects are observed directly at microscopic distances, but because the global cosmological evolution can amplify, preserve and redistribute their consequences across the observable Universe. The decisive evidence would not be a single spectacular anomaly, but a network of independent measurements converging on the same microscopic description.

11. Discussion and Outlook

The purpose of this review has been to explore a different way of approaching quantum gravity. Rather than beginning with a microscopic theory and asking how it might eventually become observable, we have started from a global property of the Universe—its possible finiteness—and asked what observable consequences follow if a finite gravitating region undergoes collapse, reaches a high-density quantum state and re-expands.
Within this perspective, the Black Hole Universe provides a concrete cosmological setting in which the problem of quantum gravity becomes connected to observations across a wide range of scales. The essential assumption is not that the Universe is a black hole, but that it is finite. Finiteness gives the system a total mass, a global gravitational radius, positive-curvature collapse dynamics and a physical boundary. Once the collapse reaches sufficiently high density, quantum physics must determine the fate of the system. A nonsingular bounce then becomes a natural arena in which microscopic quantum-gravitational effects can be transferred to the expanding Universe.
This separation between the global cosmological framework and the microscopic quantum theory is important. The BHU construction does not require one unique theory of quantum gravity. Its role is instead to provide a physical setting in which candidate theories can make observable predictions. A viable microscopic completion must determine, or at least constrain, the properties of the high-density state, the bounce dynamics, the primordial fluctuation spectrum, the symmetry structure of the quantum state and the transition from gravity to matter.
Quadratic gravity provides one particularly useful worked example. Its R 2 sector naturally generates the quasi-de Sitter dynamics associated with inflation and fixes the coupling α through the observed scalar amplitude. Its Weyl-squared sector becomes relevant in strongly inhomogeneous spacetimes and can therefore be tested through black-hole perturbations and gravitational waves. The theory consequently illustrates how cosmological and strong-gravity observations can probe different sectors of the same ultraviolet completion.
The direct-sum formulation discussed here plays a similar role for the quantum state. It provides one explicit realization of a PT-symmetric description in which complementary time orientations are incorporated into the full quantum theory. However, the observational arguments do not depend uniquely on this construction. More generally, any quantum-gravity framework in which the fundamental state possesses an approximate PT symmetry and the observed semiclassical branch breaks that symmetry can be confronted with the same parity-sensitive observables.
The emphasis should therefore remain on the observational consequences rather than on allegiance to a particular microscopic realization.

11.1. A Finite Universe as an Observational Framework

The finite cosmological setting changes the role of several familiar ingredients of early-Universe physics.
Inflation, for example, is no longer required primarily to explain why the underlying comoving spatial curvature is extremely small. In a bouncing history, the inflationary expansion is preceded by a corresponding deflationary contraction. To leading order these phases reverse one another, while the comoving curvature remains fixed. The residual curvature can therefore retain information about the initial size and binding energy of the finite perturbation that collapsed.
Likewise, primordial structure need not have a single origin. Nearly Gaussian, approximately scale-invariant fluctuations can be generated during inflation, while strongly nonlinear structures formed during contraction can survive the bounce as relic perturbations. This separation allows compact objects to originate from pre-bounce nonlinear structure without requiring the inflationary fluctuation spectrum itself to develop an extreme non-Gaussian tail.
The same global framework also connects the late-time cosmological constant to the finite boundary of the system and opens the possibility that the observed accelerated expansion reflects the asymptotic geometry of the finite gravitational region rather than an unrelated vacuum-energy component.
These examples illustrate the broader point: once the Universe is finite, global quantities that are absent or ambiguous in an infinite cosmology become physical and potentially observable.

11.2. From Phenomenology to Prediction

At present, several elements of the framework remain phenomenological. The bounce density ρ G is introduced as the characteristic density of the high-density quantum-supported state. The pre-bounce fluctuation spectrum is parametrized by quantities such as σ 8 pre and n pre . The conversion of primordial PT breaking into a B − L asymmetry is represented by an efficiency parameter ξ .
These quantities should not remain free parameters in a complete theory.
The next theoretical step is therefore to derive relations such as
ρ G = ρ G ( α , … ) ,
Y B − L = Y B − L ( A PT , α , … ) ,
and
P pre ( k ) = P pre ( k | parent background ) ,
from microscopic dynamics.
Each such relation reduces the dimensionality of the parameter space and makes the framework more predictive.
The ideal outcome would be that quantities currently inferred independently from the CMB, baryon abundance, relic populations, curvature and gravitational waves become different measurements of a much smaller set of fundamental parameters.

11.3. The Importance of Independent Observational Channels

The framework becomes compelling only if the same underlying physics is tested through genuinely independent observations.
The scalar amplitude A s constrains the inflationary gravitational sector. The parity amplitude A PT probes the symmetry of the primordial quantum state. The baryon yield Y B probes the conversion from the high-curvature state to matter. Spatial curvature probes the initial binding of the finite perturbation. Relic compact objects probe both the contracting structure spectrum and the bounce scale. Quasinormal modes, EMRIs and horizon absorption probe strong gravity in the late Universe.
These observables are separated by enormous differences in scale and cosmic time. Their value lies precisely in this independence.
A successful model must therefore satisfy a collection of relations of the schematic form
A s , A PT , Y B , Ω k , Ω relic , n SMBH , GW ⟶ one consistent microscopic description .
The objective is not to explain each observation separately. It is to over-constrain the theory.

11.4. Near-Term Tests

Several parts of this programme can be developed without waiting for direct access to Planckian physics. The relic calculation can be improved by replacing the simple pre-bounce power-law spectrum with a more realistic treatment of perturbation growth during contraction, including radiation, nonlinear evolution, mergers and the dependence on spatial curvature. Numerical simulations can then predict the full relic mass function rather than only its approximate abundance.
These predictions can be confronted with the observed dark-matter density, high-redshift black-hole populations, gravitational lensing, merger rates and stochastic gravitational-wave backgrounds.
The primordial PT interpretation can be tested through polarization. A genuine cosmological parity asymmetry should appear consistently in temperature, E- and B-mode polarization and, potentially, in tensor helicity. This provides one of the clearest near-term falsification tests of the symmetry interpretation.
The baryogenesis connection requires a microscopic calculation of reheating in the presence of PT breaking. The key question is whether a non-zero B − L yield of order
Y B − L ∼ A PT α
can emerge naturally without introducing unnaturally small or large efficiency factors.
On the strong-gravity side, future ringdown spectroscopy and EMRI observations can test the additional gravitational sectors independently of the cosmological calculations. These measurements are especially valuable because they probe the Weyl-dominated regime that is absent in an exactly homogeneous cosmological background.

11.5. What Remains Model Dependent

It is equally important to separate the broad observational strategy from claims that currently depend on specific implementations.
The identification
ρ G ∼ ? M Pl 4 8 α
is a testable hypothesis, not a derived result.
The relation
Y B − L = ξ A PT α
is a motivated phenomenological ansatz whose microscopic origin remains to be established.
Likewise, the detailed relic mass function depends on assumptions about the pre-bounce power spectrum, curvature and nonlinear evolution.
These uncertainties do not undermine the observational programme. They identify precisely where additional theoretical work is required and where future observations can discriminate among possible realizations.
The framework is therefore deliberately modular. The finite bouncing cosmology specifies the global structure; candidate quantum-gravity theories supply the microscopic dynamics; observations decide which combinations are viable.

11.6. Beyond a Single Microscopic Realization

This modularity is particularly important for the broader interpretation of the results.
Quadratic gravity is attractive because it is perturbatively renormalizable, contains a successful inflationary sector and provides additional strong-gravity degrees of freedom that can be tested observationally. PT-symmetric and direct-sum constructions are attractive because they offer possible resolutions of unitarity and horizon-related problems while predicting symmetry signatures in the primordial state.
But neither ingredient should be regarded as mandatory for the finite Universe programme.
Other quantum-gravity theories could predict a different bounce density, different primordial symmetry structure, different tensor spectrum or different strong-gravity corrections. The same observational architecture can be used to test them.
The essential requirements are instead that the theory provide a regular high-density evolution, connect contraction to expansion consistently, recover the observed semiclassical Universe and make predictions that can be compared across independent observational channels.
In this sense, the BHU should be viewed less as a complete microscopic theory than as an observational framework in which quantum gravity can be confronted with data.

11.7. Outlook

The conventional view is that quantum gravity becomes important only at microscopic distances or enormous energies that are inaccessible to experiment. A finite bouncing Universe offers a different possibility.
If the observed expansion emerged from the high-density endpoint of a previous gravitational collapse, then quantum gravity has already acted on the entire observable Universe. Its effects need not remain confined to the microscopic bounce. They can be encoded in the amplitude and symmetry of primordial fluctuations, in the abundance of matter over antimatter, in the spatial curvature, in relic compact objects and in the strong-field dynamics of black holes.
The Universe can therefore preserve information about its quantum gravitational phase across an extraordinary hierarchy of scales.
The central challenge is to determine whether these apparently different signals converge on a common microscopic description. If they do, quantum gravity may become accessible not through a single decisive experiment, but through the consistency of cosmology, particle physics and strong-gravity observations.
A finite Universe would then be more than a possible cosmological geometry. It would provide a natural laboratory for quantum gravity.

Funding

This work was partially supported by grants from Spain Plan Nacional (PGC2018-102021-B-100) and Maria de Maeztu (CEX2020-001058-M) and from European Union LACEGAL 734374 and EWC 776247. IEEC is funded by Generalitat de Catalunya.

Data Availability Statement

No new data are presented.

Acknowledgments

I would like to express my deepest gratitude to Emilio Elizalde, who supervised my PhD and has played a uniquely important role in my scientific life. He was my co-author on my first research paper, supported me at crucial stages of my career and, with great generosity, encouraged me as I developed an independent scientific path of my own. Over the years our relationship grew from that of supervisor and student into a long-standing collaboration and friendship, and we shared much of the scientific life of the institute that he helped to create. Emilio’s generosity has never been limited to science itself: he has consistently supported the careers and independence of his students and colleagues. It is therefore especially meaningful for me to contribute this review to a volume celebrating his 75th birthday, with deep gratitude for his guidance, friendship and support over more than three decades.

Conflicts of Interest

The author declares no conflict of interest.

References

  1. Rovelli, C. Quantum Gravity; Cambridge University Press, 2004. [Google Scholar]
  2. Kiefer, C. Quantum Gravity, 3 ed.; Oxford University Press, 2012. [Google Scholar]
  3. Donoghue, J.F. General Relativity as an Effective Field Theory: The Leading Quantum Corrections. Phys. Rev. D. 1994, 50, 3874–3888. [Google Scholar] [CrossRef] [PubMed]
  4. Burgess, C.P. Quantum Gravity in Everyday Life: General Relativity as an Effective Field Theory. Living Rev. Relativ. 2004, 7, 5. [Google Scholar] [CrossRef] [PubMed]
  5. Amelino-Camelia, G. Quantum-Spacetime Phenomenology. Living Rev. Relativ. 2013, 16(5), 0806.0339. [Google Scholar] [CrossRef] [PubMed]
  6. Gaztañaga, E. How the Big Bang Ends Up Inside a Black Hole. Universe 2022, 8, 257. [Google Scholar] [CrossRef]
  7. Gaztañaga, E.; Kumar, K.S.; Pradhan, S.; Gabler, M. Gravitational bounce from the quantum exclusion principle. PRD 2025, 111, 103537, [2505.23877. [Google Scholar] [CrossRef]
  8. Gaztañaga, E. The Black Hole Universe, part I. Symmetry 2022, 14, 1849. [Google Scholar] [CrossRef]
  9. Gaztañaga, E. Cosmological bounce relics: Black holes, gravitational waves, and dark matter. Phys. Rev. D. 2026, 113, 043544, [2602.17702. [Google Scholar] [CrossRef]
  10. Starobinsky, A.A. A new type of isotropic cosmological models without singularity. Phys. Lett. B 1980, 91, 99–102. [Google Scholar] [CrossRef]
  11. Sravan Kumar, K.; Marto, J. Unitary Quadratic Quantum Gravity in 4D. arXiv 2026, arXiv:2604.19707. [Google Scholar] [CrossRef]
  12. Oppenheimer, J.R.; Snyder, H. On Continued Gravitational Contraction. Phys. Rev. 1939, 56, 455–459. [Google Scholar] [CrossRef]
  13. Misner, C.W.; Sharp, D.H. Relativistic Equations for Adiabatic, Spherically Symmetric Gravitational Collapse. Phys. Rev. 1964, 136, B571–B576. [Google Scholar] [CrossRef]
  14. Gaztañaga, E. Boundary Terms, Finite FLRW Spacetimes, and the Emergence of the Cosmological Constant. Preprints.org 2026. [Google Scholar] [CrossRef]
  15. Gaztañaga, E. The mass of our observable Universe. MNRAS 2023, 521, L59–L63. [Google Scholar] [CrossRef]
  16. Elizalde, E.; Gaztañaga, E. On the probability distribution for the cosmological constant. Phys. Lett. B 1990, 234, 265–270. [Google Scholar] [CrossRef]
  17. Feynman, R.P.; Hibbs, A.R. Quantum Mechanics and Path Integrals; McGraw-Hill: New York, 1965. [Google Scholar]
  18. Gell-Mann, M.; Hartle, J.B. Quantum Mechanics in the Light of Quantum Cosmology. Complex. Entropy Phys. Inf. 1990, 8, 425–458. [Google Scholar] [CrossRef]
  19. Halliwell, J.J. Introductory Lectures on Quantum Cosmology. Quantum Cosmol. Baby Universes 1991, arXiv:gr, 159–243. [Google Scholar]
  20. Dodelson, S. Modern cosmology; Academic Press: NY, 2003. [Google Scholar]
  21. Stelle, K.S. Renormalization of Higher-Derivative Quantum Gravity. Phys. Rev. D. 1977, 16, 953. [Google Scholar] [CrossRef]
  22. Salvio, A. Quadratic Gravity. Front. Phys. 2018, 6, 77. [Google Scholar] [CrossRef]
  23. Gaztañaga, E.; Sravan Kumar, K.; Marto, J. A new understanding of Einstein–Rosen bridges. Class. Quantum Gravity 2026, 43, 015023. [Google Scholar] [CrossRef]
  24. Akrami, Y.; et al. Planck 2018 results. VII. Isotropy and Statistics of the CMB. Astron. Astrophys. 2020, 641(A7), 1906.02552. [Google Scholar] [CrossRef]
  25. Gaztañaga, E.; Kumar, K.S. Finding origins of CMB anomalies in the inflationary quantum fluctuations. JCAP 2024, 06, 001. [Google Scholar] [CrossRef]
  26. Planck Collaboration; Ade, P.A.R.; Aghanim, N.; Armitage-Caplan, C.; Arnaud, M.; Ashdown, M.; Atrio-Barand ela, F.; Aumont, J.; Baccigalupi, C.; Banday, A.J.; et al. Planck 2013 results. XXIII. Isotropy and statistics of the CMB. A A 2014, 571, A23, [1303.5083. [Google Scholar] [CrossRef]
  27. Collaboration, P. Planck 2018 results. VI. Cosmological parameters. arXiv [1807.06209]. 2018, arXiv:1807.06209p. [Google Scholar]
  28. Sakharov, A.D. Violation of CP Invariance, C Asymmetry, and Baryon Asymmetry of the Universe. JETP Lett. 1967, 5, 24–27. [Google Scholar]
  29. Riotto, A.; Trodden, M. Recent Progress in Baryogenesis. Annu. Rev. Nucl. Part. Sci. 1999, 49, 35–75. [Google Scholar] [CrossRef]
  30. Kuzmin, V.A.; Rubakov, V.A.; Shaposhnikov, M.E. On Anomalous Electroweak Baryon-Number Non-Conservation in the Early Universe. Phys. Lett. B 1985, 155, 36–42. [Google Scholar] [CrossRef]
  31. Fukugita, M.; Yanagida, T. Baryogenesis Without Grand Unification. Phys. Lett. B 1986, 174, 45–47. [Google Scholar] [CrossRef]
  32. Harvey, J.A.; Turner, M.S. Cosmological Baryon and Lepton Number in the Presence of Electroweak Fermion-Number Violation. Phys. Rev. D. 1990, 42, 3344–3349. [Google Scholar] [CrossRef] [PubMed]
  33. Buchmüller, W.; Peccei, R.D.; Yanagida, T. Leptogenesis as the Origin of Matter. Annu. Rev. Nucl. Part. Sci. 2005, 55, 311–355. [Google Scholar] [CrossRef]
  34. Peebles, P.J.E. The large-scale structure of the universe; 1980. [Google Scholar]
  35. Bernardeau, F.; Colombi, S.; Gaztañaga, E.; Scoccimarro, R. Large-Scale Structure of the Universe and Cosmological Perturbation Theory. Phys. Rep. 2002, 367, 1–248. [Google Scholar] [CrossRef]
  36. Gunn, J.E.; Gott; J. Richard, I. On the Infall of Matter into Clusters of Galaxies and Some Effects on Their Evolution. Astrophys. J. 1972, 176, 1–19. [Google Scholar] [CrossRef] [PubMed]
  37. Harada, T.; Yoo, C.M.; Kohri, K. Threshold of Primordial Black Hole Formation. Phys. Rev. D. 2013, arXiv:astro88, 084051. [Google Scholar] [CrossRef]
  38. Musco, I. Threshold for Primordial Black Holes: Dependence on the Shape of the Cosmological Perturbations. Phys. Rev. D. 2019, arXiv:astro100, 123524. [Google Scholar] [CrossRef]
  39. Elias-Llorach, P.; Gaztañaga, E. Abundance of Relic Black Holes. Preprints.org 2026. [Google Scholar] [CrossRef]
  40. Bañados, E.; Venemans, B.P.; Mazzucchelli, C.; et al. An 800-million-solar-mass black hole in a significantly neutral Universe at a redshift of 7.5. Nature 2018, 553, 473–476. [Google Scholar] [CrossRef] [PubMed]
  41. Zhu, Q.; et al. The growth of the gargantuan black holes powering high-redshift quasars and their impact on the formation of early galaxies and protoclusters. Mon. Not. R. Astron. Soc. 2024, 527, 1033–1054. [Google Scholar] [CrossRef]
  42. Shankar, F.; Salucci, P.; Granato, G.L.; De Zotti, G.; Danese, L. Super-massive Black Hole Demography: The Match between the Local and Accreted Mass Functions. In Monthly Notices of the Royal Astronomical Society; 2004; Volume 354, pp. 1020–1030. [Google Scholar] [CrossRef]
  43. Abbott, B.P.e.a. Observation of Gravitational Waves from a Binary Black Hole Merger. Phys. Rev. Lett. 2016, 116, 061102. [Google Scholar] [CrossRef] [PubMed]
  44. Berti, E.; Cardoso, V.; Starinets, A.O. Quasinormal modes of black holes and black branes. Class. Quantum Gravity 2009, 26, 163001. [Google Scholar] [CrossRef]
  45. Berti, E.; et al. Testing General Relativity with Present and Future Astrophysical Observations. Class. Quantum Gravity 2015, 32, 243001, [1501.07274. [Google Scholar] [CrossRef]
  46. Cardoso, V.; Pani, P. Testing the Nature of Dark Compact Objects: A Status Report. Living Rev. Relativ. 2019, 22(4), 1904.05363. [Google Scholar] [CrossRef]
  47. Regge, T.; Wheeler, J.A. Stability of a Schwarzschild singularity. Phys. Rev. 1957, 108, 1063–1069. [Google Scholar] [CrossRef]
  48. Zerilli, F.J. Effective potential for even-parity Regge–Wheeler gravitational perturbation equations. Phys. Rev. Lett. 1970, 24, 737–738. [Google Scholar] [CrossRef]
  49. Chandrasekhar, S. The Mathematical Theory of Black Holes; Oxford University Press, 1983. [Google Scholar]
  50. Poisson, E. Absorption of Mass and Angular Momentum by Black Holes: The Time-Domain Formalism. Phys. Rev. D. 2004, 70, 084044. [Google Scholar] [CrossRef]
  51. Brito, R.; Cardoso, V.; Pani, P. Superradiance. Lect. Notes Phys. 2015, 906, 1–237. [Google Scholar] [CrossRef]
  52. Amaro-Seoane, P.; et al. Intermediate and Extreme Mass-Ratio Inspirals—Astrophysics, Science Applications and Detection Using LISA. Class. Quantum Gravity 2007, 24, R113–R169. [Google Scholar] [CrossRef]
  53. Barack, L.; Pound, A. Self-Force and Radiation Reaction in General Relativity. Rep. Prog. Phys. 2019, 82, 016904, [1805.10385. [Google Scholar] [CrossRef] [PubMed]
Figure 1. The region inside the moving junction R ( a ) is described by the FLRW metric. The inner blue region denotes the sub-Hubble causal domain R H = 1 / H , while the yellow shell between R H and R ( a ) is still part of the FLRW interior but lies outside the instantaneous Hubble horizon. This super-Hubble region contains frozen large-scale modes, including the largest observable CMB multipoles. The moving junction R ( a ) remains timelike for finite cosmic time and asymptotically approaches the null surface R → r s = 2 G M . Outside the junction lies vacuum spacetime, which may be asymptotically Minkowski, deSitter, or Anti-deSitter. The boundary is not directly observable from inside the FLRW region, but its causal effect appears as the effective interior Λ − = 3 / r s 2 .
Figure 1. The region inside the moving junction R ( a ) is described by the FLRW metric. The inner blue region denotes the sub-Hubble causal domain R H = 1 / H , while the yellow shell between R H and R ( a ) is still part of the FLRW interior but lies outside the instantaneous Hubble horizon. This super-Hubble region contains frozen large-scale modes, including the largest observable CMB multipoles. The moving junction R ( a ) remains timelike for finite cosmic time and asymptotically approaches the null surface R → r s = 2 G M . Outside the junction lies vacuum spacetime, which may be asymptotically Minkowski, deSitter, or Anti-deSitter. The boundary is not directly observable from inside the FLRW region, but its causal effect appears as the effective interior Λ − = 3 / r s 2 .
Preprints 231015 g001
Figure 2. How a classical history emerges from quantum possibilities. In Feynman’s formulation of quantum mechanics, every possible path between A and B contributes an amplitude. When the characteristic action is comparable to ℏ (left), many alternatives can contribute significantly. As the action becomes large compared with ℏ, the phases of paths far from stationary action vary rapidly and their contributions increasingly cancel. Paths close to the stationary-action trajectory remain approximately in phase and reinforce one another, so a single classical history comes to dominate (right). The fading paths represent destructive interference, not possibilities that have ceased to exist. Later we will encounter the same transition in reverse, when classical gravitational collapse approaches a regime where quantum alternatives can no longer be ignored.
Figure 2. How a classical history emerges from quantum possibilities. In Feynman’s formulation of quantum mechanics, every possible path between A and B contributes an amplitude. When the characteristic action is comparable to ℏ (left), many alternatives can contribute significantly. As the action becomes large compared with ℏ, the phases of paths far from stationary action vary rapidly and their contributions increasingly cancel. Paths close to the stationary-action trajectory remain approximately in phase and reinforce one another, so a single classical history comes to dominate (right). The fading paths represent destructive interference, not possibilities that have ceased to exist. Later we will encounter the same transition in reverse, when classical gravitational collapse approaches a regime where quantum alternatives can no longer be ignored.
Preprints 231015 g002
Figure 4. The top panels show the anti-symmetric, T − ( n ^ ) = T ( n ^ ) − T ( − n ^ ) / 2 , component of the Planck 2018 CMB normalized temperature map ([24]) in North and South polar galactic caps. The visualization is such that the Galactic plane remains obscured along the edges. Parity conjugate positions can be identified by horizontally flipping into the opposite pole, where both caps have the same information with opposing signs: T − ( − n ^ ) = − T − ( n ^ ) . The bottom panel presents the observed T = T + + T − maps, including both the symmetric T + ( n ^ ) = T ( n ^ ) + T ( − n ^ ) / 2 and anti-symmetric components. The evident resemblance between the top and bottom panels indicates the prevalence of the anti-symmetric component.
Figure 4. The top panels show the anti-symmetric, T − ( n ^ ) = T ( n ^ ) − T ( − n ^ ) / 2 , component of the Planck 2018 CMB normalized temperature map ([24]) in North and South polar galactic caps. The visualization is such that the Galactic plane remains obscured along the edges. Parity conjugate positions can be identified by horizontally flipping into the opposite pole, where both caps have the same information with opposing signs: T − ( − n ^ ) = − T − ( n ^ ) . The bottom panel presents the observed T = T + + T − maps, including both the symmetric T + ( n ^ ) = T ( n ^ ) + T ( − n ^ ) / 2 and anti-symmetric components. The evident resemblance between the top and bottom panels indicates the prevalence of the anti-symmetric component.
Preprints 231015 g004
Figure 6. The baryron-to-photon ratio Y B during Big Bang Nucleosynthesis (BBN) can be measured from the abundace of light elements. From left to right: (1) a hot plasma of photons, neutrinos, electrons, protons, and neutrons in thermal equilibrium; (2) weak interactions freeze out near T ∼ 0.7 MeV, fixing the neutron–proton ratio; (3) at T ∼ 0.1 MeV, deuterium becomes stable and fusion begins; (4) most neutrons are captured into 4He, while trace amounts of 2H, 3He, and 7Li form.
Figure 6. The baryron-to-photon ratio Y B during Big Bang Nucleosynthesis (BBN) can be measured from the abundace of light elements. From left to right: (1) a hot plasma of photons, neutrinos, electrons, protons, and neutrons in thermal equilibrium; (2) weak interactions freeze out near T ∼ 0.7 MeV, fixing the neutron–proton ratio; (3) at T ∼ 0.1 MeV, deuterium becomes stable and fusion begins; (4) most neutrons are captured into 4He, while trace amounts of 2H, 3He, and 7Li form.
Preprints 231015 g006
Figure 8. Number density of relic black holes with masses > 10 4 M ⊙ for χ k = 50 Gpc (a) and 95 Gpc (b). Blue line represents the local SMBH number density by [42]. The white hatched region represents combination of σ 8 and ρ G for which Ω relic > Ω b . From [39].
Figure 8. Number density of relic black holes with masses > 10 4 M ⊙ for χ k = 50 Gpc (a) and 95 Gpc (b). Blue line represents the local SMBH number density by [42]. The white hatched region represents combination of σ 8 and ρ G for which Ω relic > Ω b . From [39].
Preprints 231015 g008
Figure 10. Network of cross-consistency tests in a finite bouncing cosmology. Rounded filled boxes denote observables, while square boxes denote theoretical parameters or intermediate model predictions. The CMB scalar amplitude A s fixes the R 2 coupling α , while the CMB parity signal determines A PT . These quantities enter the reheating and baryogenesis sector. The bounce density ρ G and the pre-bounce perturbation spectrum determine the relic mass function, whose different moments are constrained by the total relic density and the abundance of massive black-hole seeds. Strong-gravity observables independently probe the Weyl/spin–2 sector. Dashed arrows denote proposed microscopic or cross-consistency relations that remain to be derived.
Figure 10. Network of cross-consistency tests in a finite bouncing cosmology. Rounded filled boxes denote observables, while square boxes denote theoretical parameters or intermediate model predictions. The CMB scalar amplitude A s fixes the R 2 coupling α , while the CMB parity signal determines A PT . These quantities enter the reheating and baryogenesis sector. The bounce density ρ G and the pre-bounce perturbation spectrum determine the relic mass function, whose different moments are constrained by the total relic density and the abundance of massive black-hole seeds. Strong-gravity observables independently probe the Weyl/spin–2 sector. Dashed arrows denote proposed microscopic or cross-consistency relations that remain to be derived.
Preprints 231015 g010
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
Copyright: This open access article is published under a Creative Commons CC BY 4.0 license, which permit the free download, distribution, and reuse, provided that the author and preprint are cited in any reuse.