Cross Coherence Purity Overview

• CCP is a family of purity–coherence notions that generalizes cross-spectral purity from classical optics to diagnostic measures in quantum many-body systems.
• In optical systems, CCP ensures the invariance of normalized spectral shapes across spatial points and extends to Stokes-parameter formulations for vector fields.
• In many-body physics, CCP is defined via the squared Hilbert–Schmidt norm of reduced cross-density matrices, linking off-diagonal coherence to subsystem purity.

Searching arXiv for papers and terminology around “Cross Coherence Purity” and closely related “cross-spectral purity.” Cross Coherence Purity (CCP) denotes a family of purity–coherence notions rather than a single universally fixed definition. In statistical optics, the closest established term is cross-spectral purity, where “purity” refers to invariance of normalized spectral shape across interfering spatial points or, in vector fields, across Stokes-parameter spectra (Koivurova et al., 2024). In nonstationary electromagnetic optics, this concept has been reformulated directly in terms of the Stokes parameters so as to remain faithful to Mandel’s original scalar notion (Laatikainen et al., 2023). In a distinct quantum many-body usage, CCP is an explicitly defined quantity: the squared Hilbert–Schmidt norm of a reduced cross-density matrix between two eigenstates on a subsystem (Wang et al., 6 Mar 2026). Other recent works develop closely related coherence–purity correspondences without adopting CCP as the formal term, including basis-dependent $l^2$-coherence linked to purity, population–coherence decompositions of purity, and fidelity-based mappings between maximal coherence and purity (Soulas, 2024, Gil, 17 Feb 2026, S et al., 2021).

1. Terminological scope

The literature supports at least three technically distinct uses of the purity–coherence theme associated with CCP. In one usage, CCP is effectively a synonym for cross-spectral purity (CSP) in nonstationary optics, especially when the focus is on correlations across space and frequency or across Stokes-parameter spectra. In another, it is an explicit many-body diagnostic of cross-eigenstate coherence on a subsystem. In a third, broader and more interpretive usage, it refers to frameworks in which purity and coherence are quantitatively cross-linked but not named CCP (Joshi et al., 2023, Wang et al., 6 Mar 2026, Soulas, 2024).

Context	Object	Role
Nonstationary scalar and vector optics	Cross-spectral purity	Spectral-shape invariance across interfering points or Stokes-parameter spectra
Kinetically constrained many-body systems	$$\mathcal T_{i,i'}=\mathrm{Tr}\!\left(\rho_A^{i,i'\dagger}\rho_A^{i,i'}\right)$$	Operator-independent measure of local coherence between two eigenstates
Quantum-information and Page-type settings	CCP-like purity–coherence relations	Links between off-diagonal structure, purity, and basis or population structure

A persistent source of ambiguity is that some papers explicitly state that they do not use “Cross Coherence Purity” as a separate named concept, even when their constructions are closely related in spirit. For example, the nonstationary vector-optics paper identifies the intended concept as CSP, not CCP (Joshi et al., 2023). Likewise, the ontological coherence, black-hole evaporation, and fidelity-based purity papers all develop CCP-like structures without introducing CCP as their formal terminology (Soulas, 2024, Gil, 17 Feb 2026, S et al., 2021).

2. Optical origin: Mandel-type cross-spectral purity

In the scalar nonstationary optical setting, cross-spectral purity is formulated for a field $E(\boldsymbol{\rho};\omega)$ through the cross-spectral density

$$W(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega_1,\omega_2) = \langle E^*(\boldsymbol{\rho}_1;\omega_1)\,E(\boldsymbol{\rho}_2;\omega_2)\rangle,$$

with average spectrum

$$S(\boldsymbol{\rho};\omega)=W(\boldsymbol{\rho},\boldsymbol{\rho};\omega,\omega),$$

and normalized complex degree of coherence

$$\mu(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega_1,\omega_2) = \frac{W(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega_1,\omega_2)} {\sqrt{S(\boldsymbol{\rho}_1;\omega_1)S(\boldsymbol{\rho}_2;\omega_2)}}.$$

Mandel’s purity condition is expressed as

$$S(\boldsymbol{\rho}_1;\omega) = \frac{S(\boldsymbol{\rho}_2;\omega)}{C(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2)} = \frac{S(\boldsymbol{R};\omega)}{D(\boldsymbol{R},\boldsymbol{\rho}_1)},$$

with real, position-dependent scaling factors. The content of the condition is that normalized spectral shape is preserved across the two input points and their superposition point (Koivurova et al., 2024).

This optical framework distinguishes local from global cross-spectral purity. Local purity means Mandel’s condition holds at some special points. Global purity means it holds across the whole wavefront for all relevant spatial pairs. The global notion is the stronger one, and it is the version connected most directly to spatiotemporal separability (Koivurova et al., 2024).

The interference representation makes the physical meaning explicit. For a superposition

$$E(\boldsymbol{R};\omega) = E(\boldsymbol{\rho}_1;\omega)+E(\boldsymbol{\rho}_2;\omega)e^{i\omega\tau},$$

the equal-frequency spectrum at $\boldsymbol{R}$ is

$$\begin{aligned} S(\boldsymbol{R};\omega) &= S(\boldsymbol{\rho}_1;\omega)+S(\boldsymbol{\rho}_2;\omega) \ &\quad +2\sqrt{S(\boldsymbol{\rho}_1;\omega)S(\boldsymbol{\rho}_2;\omega)} \,|\mu(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega,\omega)| \ &\qquad \times \cos\!\left[\phi(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega,\omega)+\omega\tau\right]. \end{aligned}$$

Under complete coherence, $$\mathcal T_{i,i'}=\mathrm{Tr}\!\left(\rho_A^{i,i'\dagger}\rho_A^{i,i'}\right)$$0, and the phase structure directly governs whether the interference term can preserve spectral shape. The paper shows a two-way relation in this case: global cross-spectral purity implies spatiotemporal separability, and spatiotemporal separability implies global cross-spectral purity (Koivurova et al., 2024).

For partially coherent fields, the relevant separability notion shifts from the field itself to the correlation function. The paper therefore treats cross-spectral purity as a generalization of spatiotemporal separability to partially coherent light, but it also states that global purity does not automatically guarantee separability when the coherence function vanishes in extended regions. In the fully incoherent case, global cross-spectral purity no longer determines separability at all (Koivurova et al., 2024).

3. Stokes-parameter reformulation for random nonstationary electromagnetic beams

The vector extension begins with a two-component random electromagnetic field

$$\mathcal T_{i,i'}=\mathrm{Tr}\!\left(\rho_A^{i,i'\dagger}\rho_A^{i,i'}\right)$$1

whose cross-spectral density matrix is

$$\mathcal T_{i,i'}=\mathrm{Tr}\!\left(\rho_A^{i,i'\dagger}\rho_A^{i,i'}\right)$$2

The coherence Stokes parameters are then defined by

$$\mathcal T_{i,i'}=\mathrm{Tr}\!\left(\rho_A^{i,i'\dagger}\rho_A^{i,i'}\right)$$3

where $$\mathcal T_{i,i'}=\mathrm{Tr}\!\left(\rho_A^{i,i'\dagger}\rho_A^{i,i'}\right)$$4 is the identity and $$\mathcal T_{i,i'}=\mathrm{Tr}\!\left(\rho_A^{i,i'\dagger}\rho_A^{i,i'}\right)$$5 are the Pauli matrices. At a single space-frequency point,

$$\mathcal T_{i,i'}=\mathrm{Tr}\!\left(\rho_A^{i,i'\dagger}\rho_A^{i,i'}\right)$$6

with $$\mathcal T_{i,i'}=\mathrm{Tr}\!\left(\rho_A^{i,i'\dagger}\rho_A^{i,i'}\right)$$7 the spectral density and $$\mathcal T_{i,i'}=\mathrm{Tr}\!\left(\rho_A^{i,i'\dagger}\rho_A^{i,i'}\right)$$8 the polarization-state Stokes parameters (Laatikainen et al., 2023).

The central conceptual shift of the 2023 reformulation is that cross-spectral purity is imposed directly on the Stokes-parameter spectrum itself: $$\mathcal T_{i,i'}=\mathrm{Tr}\!\left(\rho_A^{i,i'\dagger}\rho_A^{i,i'}\right)$$9 where $E(\boldsymbol{\rho};\omega)$0 are frequency-independent constants and $E(\boldsymbol{\rho};\omega)$1 may depend on $E(\boldsymbol{\rho};\omega)$2. If this condition holds for $E(\boldsymbol{\rho};\omega)$3 with the same $E(\boldsymbol{\rho};\omega)$4, then the full polarization state is cross-spectrally pure. The paper explicitly contrasts this with earlier nonstationary vector-field treatments by noting that no extra spectral-correlation conditions are added; the reformulation is intended to remain in line with Mandel’s original definition (Laatikainen et al., 2023).

A related vector-optics paper develops the same subject under the name CSP and states the first purity condition in terms of normalized Stokes parameters,

$E(\boldsymbol{\rho};\omega)$5

with

$E(\boldsymbol{\rho};\omega)$6

That work emphasizes that nonstationary vector fields do not admit the stationary-style reduction formula directly at the level of raw normalized two-time Stokes parameters; instead, a time-integrated formulation is required (Joshi et al., 2023).

4. Reduction formulas and constrained correlation structure

The Stokes-parameter reformulation introduces the time-integrated mutual coherence matrix

$E(\boldsymbol{\rho};\omega)$7

and the associated coherence Stokes parameters

$E(\boldsymbol{\rho};\omega)$8

The preferred normalized object is the intensity-normalized coherence Stokes parameter

$E(\boldsymbol{\rho};\omega)$9

If $$W(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega_1,\omega_2) = \langle E^*(\boldsymbol{\rho}_1;\omega_1)\,E(\boldsymbol{\rho}_2;\omega_2)\rangle,$$0 is cross-spectrally pure, then

$$W(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega_1,\omega_2) = \langle E^*(\boldsymbol{\rho}_1;\omega_1)\,E(\boldsymbol{\rho}_2;\omega_2)\rangle,$$1

This reduction formula separates the normalized time-integrated two-point coherence into spatial, temporal, and—when $$W(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega_1,\omega_2) = \langle E^*(\boldsymbol{\rho}_1;\omega_1)\,E(\boldsymbol{\rho}_2;\omega_2)\rangle,$$2—polarization contributions (Laatikainen et al., 2023).

The polarization-dependent factor is

$$W(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega_1,\omega_2) = \langle E^*(\boldsymbol{\rho}_1;\omega_1)\,E(\boldsymbol{\rho}_2;\omega_2)\rangle,$$3

where

$$W(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega_1,\omega_2) = \langle E^*(\boldsymbol{\rho}_1;\omega_1)\,E(\boldsymbol{\rho}_2;\omega_2)\rangle,$$4

For $$W(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega_1,\omega_2) = \langle E^*(\boldsymbol{\rho}_1;\omega_1)\,E(\boldsymbol{\rho}_2;\omega_2)\rangle,$$5, the reduction contains only the usual spatial and temporal factors. For $$W(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega_1,\omega_2) = \langle E^*(\boldsymbol{\rho}_1;\omega_1)\,E(\boldsymbol{\rho}_2;\omega_2)\rangle,$$6, the extra factor is a polarization-state average. The paper identifies this third factor as a direct consequence of using intensity-normalized coherence Stokes parameters (Laatikainen et al., 2023).

The same reformulation imposes a specific structure on spectral spatial correlations. With

$$W(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega_1,\omega_2) = \langle E^*(\boldsymbol{\rho}_1;\omega_1)\,E(\boldsymbol{\rho}_2;\omega_2)\rangle,$$7

cross-spectral purity yields, at equal frequencies,

$$W(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega_1,\omega_2) = \langle E^*(\boldsymbol{\rho}_1;\omega_1)\,E(\boldsymbol{\rho}_2;\omega_2)\rangle,$$8

Hence $$W(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega_1,\omega_2) = \langle E^*(\boldsymbol{\rho}_1;\omega_1)\,E(\boldsymbol{\rho}_2;\omega_2)\rangle,$$9 is frequency independent, as in scalar CCP theory, whereas for $$S(\boldsymbol{\rho};\omega)=W(\boldsymbol{\rho},\boldsymbol{\rho};\omega,\omega),$$0 the frequency dependence is determined only by the local spectral polarization states at the two points (Laatikainen et al., 2023).

The earlier nonstationary vector paper derives a parallel time-integrated reduction law,

$$S(\boldsymbol{\rho};\omega)=W(\boldsymbol{\rho},\boldsymbol{\rho};\omega,\omega),$$1

and extends strict CSP through matched space-frequency and time-integrated space-time quantities, including the electromagnetic degree of coherence and the degree of circular polarization. Its central methodological point is that the stationary-field reduction structure is recovered only after time integration (Joshi et al., 2023).

5. Explicit CCP in kinetically constrained many-body systems

A different and explicit use of the term appears in the study of quantum many-body scars. There CCP is introduced for a local observable $$S(\boldsymbol{\rho};\omega)=W(\boldsymbol{\rho},\boldsymbol{\rho};\omega,\omega),$$2 supported on a subsystem $$S(\boldsymbol{\rho};\omega)=W(\boldsymbol{\rho},\boldsymbol{\rho};\omega,\omega),$$3, with off-diagonal matrix elements written via the reduced cross-density matrix

$$S(\boldsymbol{\rho};\omega)=W(\boldsymbol{\rho},\boldsymbol{\rho};\omega,\omega),$$4

The cross coherence purity is defined as

$$S(\boldsymbol{\rho};\omega)=W(\boldsymbol{\rho},\boldsymbol{\rho};\omega,\omega),$$5

This is the squared Hilbert–Schmidt norm of the reduced cross-density matrix, and it is introduced as an operator-independent measure of how strongly two many-body eigenstates remain coherently connected after restricting to a local subsystem (Wang et al., 6 Mar 2026).

The quantity has several structural properties. It is symmetric,

$$S(\boldsymbol{\rho};\omega)=W(\boldsymbol{\rho},\boldsymbol{\rho};\omega,\omega),$$6

because $$S(\boldsymbol{\rho};\omega)=W(\boldsymbol{\rho},\boldsymbol{\rho};\omega,\omega),$$7. In the diagonal limit,

$$S(\boldsymbol{\rho};\omega)=W(\boldsymbol{\rho},\boldsymbol{\rho};\omega,\omega),$$8

so CCP reduces to the usual subsystem purity, i.e. the second Rényi purity. The paper also states that $$S(\boldsymbol{\rho};\omega)=W(\boldsymbol{\rho},\boldsymbol{\rho};\omega,\omega),$$9 depends on the subsystem basis used to define $$\mu(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega_1,\omega_2) = \frac{W(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega_1,\omega_2)} {\sqrt{S(\boldsymbol{\rho}_1;\omega_1)S(\boldsymbol{\rho}_2;\omega_2)}}.$$0, but CCP itself is operator-independent once $$\mu(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega_1,\omega_2) = \frac{W(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega_1,\omega_2)} {\sqrt{S(\boldsymbol{\rho}_1;\omega_1)S(\boldsymbol{\rho}_2;\omega_2)}}.$$1 is fixed (Wang et al., 6 Mar 2026).

CCP is embedded in a revised ETH framework in which eigenstates are characterized not only by energy but also by a quasiparticle number

$$\mu(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega_1,\omega_2) = \frac{W(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega_1,\omega_2)} {\sqrt{S(\boldsymbol{\rho}_1;\omega_1)S(\boldsymbol{\rho}_2;\omega_2)}}.$$2

The generalized density of states becomes $$\mu(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega_1,\omega_2) = \frac{W(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega_1,\omega_2)} {\sqrt{S(\boldsymbol{\rho}_1;\omega_1)S(\boldsymbol{\rho}_2;\omega_2)}}.$$3, and the proposed off-diagonal scaling implies

$$\mu(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega_1,\omega_2) = \frac{W(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega_1,\omega_2)} {\sqrt{S(\boldsymbol{\rho}_1;\omega_1)S(\boldsymbol{\rho}_2;\omega_2)}}.$$4

Numerically, low-$$\mu(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega_1,\omega_2) = \frac{W(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega_1,\omega_2)} {\sqrt{S(\boldsymbol{\rho}_1;\omega_1)S(\boldsymbol{\rho}_2;\omega_2)}}.$$5 regions exhibit enhanced CCP. In this framework, generic high-density regions have suppressed off-diagonal coherence, whereas scarred low-density regions retain large coherence, producing anomalously large temporal fluctuations and quasi-periodic revivals (Wang et al., 6 Mar 2026).

6. Related quantum frameworks linking coherence and purity

Several papers develop structures that are closely related to CCP while using different terminology. One such framework reinterprets coherence as deviation from the total probability formula and defines a coherence measure as a function of both a state and a basis,

$$\mu(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega_1,\omega_2) = \frac{W(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega_1,\omega_2)} {\sqrt{S(\boldsymbol{\rho}_1;\omega_1)S(\boldsymbol{\rho}_2;\omega_2)}}.$$6

Among the measures satisfying the paper’s axioms are

$$\mu(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega_1,\omega_2) = \frac{W(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega_1,\omega_2)} {\sqrt{S(\boldsymbol{\rho}_1;\omega_1)S(\boldsymbol{\rho}_2;\omega_2)}}.$$7

and

$$\mu(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega_1,\omega_2) = \frac{W(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega_1,\omega_2)} {\sqrt{S(\boldsymbol{\rho}_1;\omega_1)S(\boldsymbol{\rho}_2;\omega_2)}}.$$8

Its main purity-related result states that, for a Haar-random orthonormal basis,

$$\mu(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega_1,\omega_2) = \frac{W(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2;\omega_1,\omega_2)} {\sqrt{S(\boldsymbol{\rho}_1;\omega_1)S(\boldsymbol{\rho}_2;\omega_2)}}.$$9

Accordingly, $$S(\boldsymbol{\rho}_1;\omega) = \frac{S(\boldsymbol{\rho}_2;\omega)}{C(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2)} = \frac{S(\boldsymbol{R};\omega)}{D(\boldsymbol{R},\boldsymbol{\rho}_1)},$$0 is interpreted as the approximate level of $$S(\boldsymbol{\rho}_1;\omega) = \frac{S(\boldsymbol{\rho}_2;\omega)}{C(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2)} = \frac{S(\boldsymbol{R};\omega)}{D(\boldsymbol{R},\boldsymbol{\rho}_1)},$$1-coherence in most bases, asymptotically and in the Haar-average sense (Soulas, 2024).

A second framework decomposes normalized global purity into population and coherence contributions. For an $$S(\boldsymbol{\rho}_1;\omega) = \frac{S(\boldsymbol{\rho}_2;\omega)}{C(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2)} = \frac{S(\boldsymbol{R};\omega)}{D(\boldsymbol{R},\boldsymbol{\rho}_1)},$$2-dimensional density matrix,

$$S(\boldsymbol{\rho}_1;\omega) = \frac{S(\boldsymbol{\rho}_2;\omega)}{C(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2)} = \frac{S(\boldsymbol{R};\omega)}{D(\boldsymbol{R},\boldsymbol{\rho}_1)},$$3

with complementary indices

$$S(\boldsymbol{\rho}_1;\omega) = \frac{S(\boldsymbol{\rho}_2;\omega)}{C(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2)} = \frac{S(\boldsymbol{R};\omega)}{D(\boldsymbol{R},\boldsymbol{\rho}_1)},$$4

satisfying

$$S(\boldsymbol{\rho}_1;\omega) = \frac{S(\boldsymbol{\rho}_2;\omega)}{C(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2)} = \frac{S(\boldsymbol{R};\omega)}{D(\boldsymbol{R},\boldsymbol{\rho}_1)},$$5

In Page-type evaporation, if radiation populations remain nearly uniform in the chosen basis, then $$S(\boldsymbol{\rho}_1;\omega) = \frac{S(\boldsymbol{\rho}_2;\omega)}{C(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2)} = \frac{S(\boldsymbol{R};\omega)}{D(\boldsymbol{R},\boldsymbol{\rho}_1)},$$6 and late-time purity recovery must be coherence-dominated. This is not named CCP in the paper, but it is an explicit purity–coherence decomposition with a geometric “routes to purity” interpretation (Gil, 17 Feb 2026).

A third framework defines fidelity-based purity and coherence using

$$S(\boldsymbol{\rho}_1;\omega) = \frac{S(\boldsymbol{\rho}_2;\omega)}{C(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2)} = \frac{S(\boldsymbol{R};\omega)}{D(\boldsymbol{R},\boldsymbol{\rho}_1)},$$7

Purity is

$$S(\boldsymbol{\rho}_1;\omega) = \frac{S(\boldsymbol{\rho}_2;\omega)}{C(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2)} = \frac{S(\boldsymbol{R};\omega)}{D(\boldsymbol{R},\boldsymbol{\rho}_1)},$$8

and coherence is

$$S(\boldsymbol{\rho}_1;\omega) = \frac{S(\boldsymbol{\rho}_2;\omega)}{C(\boldsymbol{\rho}_1,\boldsymbol{\rho}_2)} = \frac{S(\boldsymbol{R};\omega)}{D(\boldsymbol{R},\boldsymbol{\rho}_1)},$$9

The paper proves that maximal coherence and purity are one-to-one related: $$E(\boldsymbol{R};\omega) = E(\boldsymbol{\rho}_1;\omega)+E(\boldsymbol{\rho}_2;\omega)e^{i\omega\tau},$$0 This again does not define CCP, but it gives a direct purity–coherence map through a common fidelity functional (S et al., 2021).

7. Conceptual boundaries and common misconceptions

One recurring misconception is to read “purity” in the optical literature as a statement merely about bandwidth or monochromaticity. The cited works use purity in a much more specific sense: the normalized spectral shape must be preserved across space, or across the relevant Stokes-parameter spectra, up to frequency-independent scaling factors (Koivurova et al., 2024, Laatikainen et al., 2023).

A second misconception is to treat local and global purity as interchangeable. The scalar field analysis distinguishes them sharply: local cross-spectral purity may hold at selected points, whereas global purity requires consistency across the whole wavefront. Only the global notion supports the strongest link to spatiotemporal separability, and even then only under the fully coherent assumptions stated in the paper (Koivurova et al., 2024).

A third source of confusion arises in partially coherent and fully incoherent regimes. The optical literature does not claim that cross-spectral purity universally determines separability. Rather, for partially coherent fields the relevant object is often the separability of the correlation function, and if the coherence function vanishes in extended regions then purity alone cannot reveal hidden spatiotemporal coupling. In the fully incoherent case, the link breaks down entirely (Koivurova et al., 2024).

In the vector nonstationary setting, the 2023 Stokes-parameter reformulation also corrects an earlier conceptual tendency to supplement purity with extra spectral-correlation constraints. Its stated purpose is precisely to impose the purity condition directly on the Stokes-parameter spectrum and thereby remain in line with Mandel’s original scalar definition. The additional polarization factor in the reduction formula is not an ad hoc insertion; it arises from the use of intensity-normalized coherence Stokes parameters (Laatikainen et al., 2023).

In many-body scar physics, CCP should likewise not be confused with the purity of a single state. The defining object is a cross quantity between two eigenstates after subsystem restriction, and ordinary subsystem purity appears only in the diagonal limit $$E(\boldsymbol{R};\omega) = E(\boldsymbol{\rho}_1;\omega)+E(\boldsymbol{\rho}_2;\omega)e^{i\omega\tau},$$1. This makes CCP a diagnostic of local inter-eigenstate coherence rather than a synonym for mixedness or Rényi purity alone (Wang et al., 6 Mar 2026).

Taken together, these works suggest a broad but non-unified landscape. In optics, CCP is best read as cross-spectral purity and its Stokes-parameter generalizations. In kinetically constrained many-body systems, CCP is a specific subsystem cross-density invariant. In adjacent quantum-information settings, the same phrase is more plausibly interpreted as a family resemblance among constructions that tie coherence to purity through basis dependence, off-diagonal weight, or population–coherence decomposition (Joshi et al., 2023, Soulas, 2024, Gil, 17 Feb 2026, S et al., 2021).