The copycat process is the transient stage between the onset of decoherence and completed einselection in which the second eigenstate of a decohering qubit's reduced density matrix takes the mirror image of the initial superposition, while the eigenvalues grow only quadratically in time. For the reduced Caldeira--Leggett model it is analytically characterized in only two regimes: a spectrum-agnostic short-time perturbative expansion, and a closed-form non-perturbative solution that requires the two conditional environmental Hamiltonians to share an eigenbasis and be proportional. The open problem is the missing generality: (a) a characterization of the copycat regime for general environmental operator pairs, including non-commuting conditional Hamiltonians with non-proportional spectra -- whether the copycat regime persists, what replaces the closed-form decoherence factor, and which features of the relative operator structure control its presence, form, and breakdown; and (b) quantitative bounds on the duration of the perturbative copycat window and of the broader copycat regime, in terms of environmental spectral properties (such as level density, correlations, and level-spacing statistics) and coupling strength, valid for general environments.

Background

When a quantum system interacts with its environment, entanglement builds up and suppresses the off-diagonal elements of the system's reduced density matrix -- the process of decoherence. The same interaction selects a preferred set of stable 'pointer states' into which superpositions decay, a mechanism called einselection (environment-induced superselection), reviewed by Zurek (Rev. Mod. Phys. 75, 715 (2003)). A standard laboratory for these questions is the Caldeira--Leggett model of a harmonic oscillator coupled to an oscillator bath. Albrecht, Baunach, and Arrasmith introduced a finite-dimensional, fully unitary 'adapted Caldeira--Leggett' (ACL) model for numerical study, together with a 'reduced Caldeira--Leggett' (RCL) limit in which a two-state system a∣↑⟩+b∣↓⟩a|\uparrow\rangle + b|\downarrow\rangle (a qubit standing for a superposition of two pointer states) couples to an environment through Hw=λ(∣↑⟩⟨↑∣⊗He↑+∣↓⟩⟨↓∣⊗He↓),H_w = \lambda\left(|\uparrow\rangle\langle\uparrow|\otimes H_e^{\uparrow} + |\downarrow\rangle\langle\downarrow|\otimes H_e^{\downarrow}\right), where He↑H_e^{\uparrow} and He↓H_e^{\downarrow} act only on the environment and need not commute.

Numerically this model family displays a striking early-time feature: shortly after decoherence begins but long before einselection is complete, the second eigenstate of the system density matrix ρs\rho_s takes the 'mirror image' of the initial superposition (amplitudes swapped and conjugated) and is transiently stable, while the eigenvalues depart from their initial values only quadratically in time. Baunach, Albrecht, and Arrasmith call this the copycat process. Expanding the unitary evolution to second order in the short time parameter Δ\Delta, they obtained ρs\rho_s with off-diagonal element ab∗(1+iβΔ−ηΔ2)ab^{*}(1+i\beta\Delta-\eta\Delta^{2}), where β\beta is real and η\eta complex environmental expectation values; eigenvalues p1=1−∣a∣2∣b∣2ϵΔ2p_{1}=1-|a|^{2}|b|^{2}\epsilon\Delta^{2} and p2=∣a∣2∣b∣2ϵΔ2p_{2}=|a|^{2}|b|^{2}\epsilon\Delta^{2}, where ϵ\epsilon is the variance of (λ/ℏ)(He↓−He↑)(\lambda/\hbar)(H_e^{\downarrow}-H_e^{\uparrow}) in the initial environmental state; and the second eigenstate in the copycat mirror form. These perturbative results hold for non-commuting He↑H_e^{\uparrow} and He↓H_e^{\downarrow} and are agnostic about the environmental spectrum. Early-time quadratic decay of coherence is a familiar short-time (quantum-Zeno-related) phenomenon; what the copycat analysis adds is the eigenstate-resolved picture of which state carries the emerging classical mixture and how it moves before einselection completes.

To reach beyond short times, the same authors specialize to environments in which He↑H_e^{\uparrow} and He↓H_e^{\downarrow} share an eigenbasis and are proportional, He↓=f↑↓He↑H_e^{\downarrow}=f^{\uparrow\downarrow}H_e^{\uparrow} (their Eqns. 34--36), an assumption they describe as what enables the analysis that follows; this yields a closed-form non-perturbative decoherence factor z(t)z(t) (a weighted sum of environmental phases) controlling the off-diagonal elements at all times. Comparison with numerics for representative random spectra shows that the perturbative regime 'can be a sizable portion of the full time evolution' of the system's linear entropy and spin expectation ⟨Sx⟩\langle S_x\rangle, but its duration is only remarked to vary somewhat with coupling strength and environmental frequency spectra, and is never quantified. Residual oscillatory coherence in the spin expectation ⟨Sx⟩\langle S_x\rangle is interpreted in the companion ACL paper as Loschmidt-echo (partial time reversal) and nuclear-magnetic-resonance (NMR)-like finite-environment effects, in the spirit of Cucchietti--Paz--Zurek's spin-environment analysis.

The unfinished program is the generality of these statements: what happens to the copycat regime when the environmental operators are not simultaneously diagonalizable or not proportional, and how long the quadratic copycat window lasts as a function of the environment's spectral properties and the coupling strength.

Problem Statement

Within the reduced Caldeira-Leggett framework -- a two-state system a∣↑⟩+b∣↓⟩a|\uparrow\rangle + b|\downarrow\rangle coupled to an environment through Hw=λ(∣↑⟩⟨↑∣⊗He↑+∣↓⟩⟨↓∣⊗He↓)H_w=\lambda(|\uparrow\rangle\langle\uparrow|\otimes H_e^{\uparrow}+|\downarrow\rangle\langle\downarrow|\otimes H_e^{\downarrow}), with the system initially unentangled from the environment -- characterize the early-time copycat regime of decoherence and einselection for general environmental operator structures, in two jointly linked parts:

(a) Behavior beyond the simplifying structural assumptions: the closed-form non-perturbative solution for the decoherence factor z(t)z(t), the eigenvalues, and the eigenstate trajectories of the system density matrix is known only when He↑H_e^{\uparrow} and He↓H_e^{\downarrow} share an eigenbasis and satisfy He↓=f↑↓He↑H_e^{\downarrow}=f^{\uparrow\downarrow}H_e^{\uparrow}. Determine how the copycat regime behaves when these assumptions are dropped -- for non-commuting conditional Hamiltonians and general (non-proportional) spectral relationships: does the copycat regime (the transient mirror-image second eigenstate with t2t^{2} eigenvalue growth) persist beyond the perturbative window, what replaces the closed-form non-perturbative decoherence factor, and which quantitative features of the relative operator structure (such as the commutator [He↑,He↓][H_e^{\uparrow},H_e^{\downarrow}], spectral overlap, or eigenbasis misalignment) control its presence, form, and breakdown? These three points are jointly required to resolve (a); addressing only one of them does not resolve it.

(b) Duration of the perturbative window: derive quantitative bounds -- as functions of environmental spectral properties (such as level density, correlations, and level-spacing statistics) and of the coupling strength λ\lambda -- on the time interval over which the O(Δ2)O(\Delta^{2}) perturbative copycat behavior dominates, and on the duration of the broader copycat regime, valid for general environments rather than representative numerical spectra. Because the end of the window is not fixed by the model, the solver must also state explicitly and justify the criterion -- for example, a tolerance on the departure of the eigenvalues from the quadratic law, or on the overlap of the second eigenstate with the mirror form -- that defines the end of each bounded interval.

The common objective of (a) and (b) is to delimit when, and for how long, the copycat process governs the earliest stage of einselection as a function of the environment's operator and spectral structure; a resolution must make the dependence on that structure explicit enough that a qualified expert can decide, for a specified environment, whether the copycat regime occurs and how long it lasts.

References

  1. Rose Baunach, Andreas Albrecht, and Andrew Arrasmith, 'Copycat process in the early stages of einselection', Phys. Rev. Research 5, 023188 (2023). DOI: 10.1103/PhysRevResearch.5.023188; arXiv:2105.14032; https://arxiv.org/abs/2105.14032
  2. Andreas Albrecht, Rose Baunach, and Andrew Arrasmith, 'Adapted Caldeira-Leggett model', Phys. Rev. Research 5, 023187 (2023). DOI: 10.1103/PhysRevResearch.5.023187; arXiv:2105.14040; https://arxiv.org/abs/2105.14040
  3. Bin Yan and Wojciech H. Zurek, 'Decoherence factor as a convolution: an interplay between a Gaussian and an exponential coherence loss', New J. Phys. 24, 113029 (2022). DOI: 10.1088/1367-2630/ac9fe8; arXiv:2110.09463; https://arxiv.org/abs/2110.09463
  4. Andreas Albrecht, 'Equilibration and "Thermalization" in the Adapted Caldeira-Leggett Model', Entropy 24, 316 (2022). DOI: 10.3390/e24030316; arXiv:2202.12353; https://arxiv.org/abs/2202.12353
  5. Luciano Manara, Andrea Smirne, and Bassano Vacchini, 'Non-Markovianity in the Adapted Caldeira-Leggett model', Int. J. Quantum Inf. 24, 2640004 (2026). DOI: 10.1142/S0219749926400046; arXiv:2605.19753; https://arxiv.org/abs/2605.19753
  6. Wojciech H. Zurek, 'Decoherence, einselection, and the quantum origins of the classical', Rev. Mod. Phys. 75, 715 (2003). DOI: 10.1103/RevModPhys.75.715; arXiv:quant-ph/0105127; https://arxiv.org/abs/quant-ph/0105127
  7. F. M. Cucchietti, J. P. Paz, and W. H. Zurek, 'Decoherence from spin environments', Phys. Rev. A 72, 052113 (2005). DOI: 10.1103/PhysRevA.72.052113; https://doi.org/10.1103/PhysRevA.72.052113
  8. Rose Baunach, 'Two Windows on the Quantum World', PhD dissertation, University of California, Davis (2023). INSPIRE literature record 2757266; https://inspirehep.net/literature/2757266