Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

arXiv:2602.19993

GAP Measures and Wave Function Collapse

Roderich Tumulka

quant-phmath-phSubmitted Feb 23, 2026• Announced Feb 24, 2026• 1 citation• Revised Nov 2, 2026 (v2)
v1 · Feb 23, 2026v2 · Nov 2, 2026
arXivHTMLPDF

Abstract

GAP measures (also known as Scrooge measures) are a natural class of probability distributions on the unit sphere of a Hilbert space that come up in quantum statistical mechanics; for each density matrix $ρ$ there is a unique measure GAP$_ρ$. We describe and prove a property of these measures that was not recognized so far: If a wave function $Ψ$ is GAP$_ρ$ distributed and a collapse occurs, then the collapsed wave function $Ψ'$ is again GAP distributed (relative to the appropriate $ρ'$). This fact applies to collapses due to a quantum measurement carried out by an observer, as well as to spontaneous collapse theories such as CSL or GRW. More precisely, it is the conditional distribution of $Ψ'$, given the measurement outcome (respectively, the noise in CSL or the collapse history in GRW), that is GAP$_{ρ'}$.

Comments
12 pages LaTeX, no figures; changes in v2: Section 4 added and minor corrections and improvements throughout the paper

Related papers

Non-Markovian collapse models

Luca Ferialdi

We introduce the measurement problem in quantum mechanics and we briefly \ndiscuss \nthe solutions proposed in literature. We then focus our \nattention on models of spontaneous wavefunction collapse. \nWe describe the two most popular models (GRW, CSL) and list other proposals. \nWe analyze \nin detail a third collapse \nmodel (QMUPL), which is particularly simple (but physically meaningful) to \nbe studied in great mathematical detail. We discuss its main properties. We \nalso describe a "finite \ntemperature" version of this model, which includes dissipative terms. \n \nThese models are Markovian, i.e. the collapse mechanism is driven by a \nwhite noise. Since the ultimate goal is to identify the noise responsible \nfor the \ncollapse with a random field in Nature, it becomes important to study \nnon-Markovian generalizations of collapse models, where the collapsing field \nhas a generic correlation function, likely with a cut off at high \nfrequencies. \n \nModels of this kind have already been studied, as a generalization of the \nCSL model. In this thesis \nwe describe in mathematical detail the generalization of the QMUPL model to \nnon-Markovian noises. After having proved, under suitable conditions, the \nseparation of the center-of-mass and relative \nmotions for a generic ensemble of particles, we focus our analysis on the \ntime evolution of the center of mass \nof an isolated system (free particle case). We compute the explicit \nexpression of the Green's function via the path integral formalism, for a \ngeneric Gaussian noise. We analyze in detail the case of an exponential \ncorrelation function, providing the exact analytical solution. \n \nWe next study the \ntime evolution of average quantities, such as the mean position, momentum \n(which satisfy Ehrefest's theorem) and energy (which is not conserved like \nin the other collapse models). We also compute the non-Markovian master \nequation \nfor an harmonic oscillator, according to this model, and compare \nits structure to the well-known Lindblad structure of Markovian open quantum \nsystems. \n \nWe \neventually specialize to the case of \nGaussian wave functions, and prove that all basic facts about collapse \nmodels (reduction process, amplification mechanism, etc.), which are known \nto be true in the white noise case, hold also in \nthe more general case of non-Markovian dynamics. \n \n \nWe further analyze the evolution \nof Gaussian wave function according to the three different realizations of \nthe QMUPL model so far developed (Markovian, non-Markovian and "finite \ntemperature"), comparing their fundamental features. \nFinally, by analyzing different localization criteria, we set new lower \nbounds on the \nparameters of these models, and we compare them with the upper bounds coming \nfrom known experimental data.

•Mar 26, 2010•0 citations

Can one detect whether a wave function has collapsed?

Charles Wesley Cowan, Roderich Tumulka

Consider a quantum system prepared in state ψ, a unit vector in a d -dimensional Hilbert space. Let b 1 , ..., b d be an orthonormal basis and suppose that, with some probability 0 < p < 1, ψ 'collapses', i.e., gets replaced by b k (possibly times a phase factor) with Born's probability |〈 b k |ψ〉| 2 . The question we investigate is: how well can any quantum experiment on the system determine afterwards whether a collapse has occurred? The answer depends on how much is known about the initial vector ψ. We provide a number of different results addressing several variants of the question. In each case, no experiment can provide more than rather limited probabilistic information. In case ψ is drawn randomly with uniform distribution over the unit sphere in Hilbert space, no experiment performs better than a blind guess without measurement; that is, no experiment provides any useful information.

arXiv:1307.0810•Journal of Physics A Mathematical and Theoretical•Apr 23, 2014•4 citations

Relation Between Quantum Jump and Wave Function Collapse

Tadashi Nakajima

Whether wave function collapses or not is a major remaining question in the theory of quantum measurement. This difficulty stems from following two facts. First, it has not been recognized that single-particle quantum mechanics and many-particle quantum mechanics must be treated separately. Second, quantum jump (QJ) and wave function collapse (WFC) need clearer definitions. We define a QJ as a process of selecting a set of system eigenvalues (SEVs) of an observable and a WFC as a process of determining the probability distribution (PD) of SEVs, both from a single measurement. The goal of quantum observation is to obtain the PD, which is determined from an ensemble of SEVs. The wave function becomes an observable when the PD is determined. In single-particle quantum mechanics, a single measurement results in only one set of SEVs and the PD is not observable. Therefore the WFC does not happen. In many-particle quantum mechanics, we focus on the occupation number of a singe quantum state. The wave function does not collapse in general, but there are exceptions. The occupation number can be huge and macroscopic for photons or for Bose-Einstein condensates. In such a case, the PD is determined from a single measurement of a real ensemble and the WFC occurs. We call it a macroscopic quantum jump, which effectively is a measurement of a classical observable.

Qeios•Sep 5, 2024•0 citations

Relation between Quantum Jump and Wave Function Collapse

Tadashi Nakajima

Whether the wave function collapses or not is a major remaining question in the theory of quantum measurement. This difficulty stems from the following two facts. First, it has not been recognized that single-particle quantum mechanics and many-particle quantum mechanics must be treated separately. Second, quantum jump (QJ) and wave function collapse (WFC) need clearer definitions. We define a QJ as a process of selecting a set of system eigenvalues (SEVs) of an observable and a WFC as a process of determining the probability distribution (PD) of SEVs, both from a single measurement. The goal of quantum observation is to obtain the PD, which is determined from an ensemble of SEVs. The wave function becomes an observable when the PD is determined. In single-particle quantum mechanics, a single measurement results in only one set of SEVs, and the PD is not observable. Therefore, the WFC does not happen. In many-particle quantum mechanics, we focus on the occupation number of a single quantum state. The wave function does not collapse in general, but there are exceptions. The occupation number can be huge and macroscopic for photons or for Bose-Einstein condensates. In such a case, the PD is determined from a single measurement of a real ensemble, and the WFC occurs. We call it a macroscopic quantum jump, which effectively is a measurement of a classical observable.

Qeios•Dec 8, 2024•9 citations

Detecting wave function collapse without prior knowledge

Charles Wesley Cowan, Roderich Tumulka

We are concerned with the problem of detecting with high probability whether a wave function has collapsed or not, in the following framework: A quantum system with a d-dimensional Hilbert space is initially in state ψ; with probability 0 < p < 1, the state collapses relative to the orthonormal basis b1, …, bd. That is, the final state ψ′ is random, it is ψ with probability 1 − p and bk (up to a phase) with p times Born’s probability bk|ψ2. Now an experiment on the system in state ψ′ is desired that provides information about whether or not a collapse has occurred. Elsewhere [C. W. Cowan and R. Tumulka, J. Phys. A: Math. Theor. 47, 195303 (2014)], we identify and discuss the optimal experiment in case that ψ is either known or random with a known probability distribution. Here, we present results about the case that no a priori information about ψ is available, while we regard p and b1, …, bd as known. For certain values of p, we show that the set of ψs for which any experiment E is more reliable than blind guessing is at most half the unit sphere; thus, in this regime, any experiment is of questionable use, if any at all. Remarkably, however, there are other values of p and experiments E such that the set of ψs for which E is more reliable than blind guessing has measure greater than half the sphere, though with a conjectured maximum of 64% of the sphere.

arXiv:1312.7321•Journal of Mathematical Physics•Aug 1, 2015•2 citations

Relation between Quantum Jump and Wave Function Collapse

Tadashi Nakajima

Whether the wave function collapses or not is a major remaining question in the theory of quantum measurement. This difficulty stems from the following two facts. First, it has not been recognized that single-particle quantum mechanics and many-particle quantum mechanics must be treated separately. Second, quantum jump (QJ) and wave function collapse (WFC) need clearer definitions. We define a QJ as a process of selecting a set of system eigenvalues (SEVs) of an observable and a WFC as a process of determining the probability distribution (PD) of SEVs, both from a single measurement. The goal of quantum observation is to obtain the PD, which is determined from an ensemble of SEVs. The wave function becomes an observable when the PD is determined. In single-particle quantum mechanics, a single measurement results in only one set of SEVs, and the PD is not observable. Therefore, the WFC does not happen. In many-particle quantum mechanics, we focus on the occupation number of a single quantum state. The wave function does not collapse in general, but there are exceptions. The occupation number can be huge and macroscopic for photons or for Bose-Einstein condensates. In such a case, the PD is determined from a single measurement of a real ensemble, and the WFC occurs. We call it a macroscopic quantum jump, which effectively is a measurement of a classical observable.

Qeios•Oct 18, 2024•0 citations
GAP measures (also known as Scrooge measures) are a natural class of probability distributions on the unit sphere of a Hilbert space that come up in quantum statistical mechanics; for each density matrix ρρρ there is a unique measure GAPρ_ρρ​. We describe and prove a property of these measures that was not recognized so far: If a wave function ΨΨΨ is GAPρ_ρρ​ distributed and a collapse occurs, then the collapsed wave function Ψ′Ψ'Ψ′ is again GAP distributed (relative to the appropriate ρ′ρ'ρ′). This fact applies to collapses due to a quantum measurement carried out by an observer, as well as to spontaneous collapse theories such as CSL or GRW. More precisely, it is the conditional distribution of Ψ′Ψ'Ψ′, given the measurement outcome (respectively, the noise in CSL or the collapse history in GRW), that is GAPρ′_{ρ'}ρ′​.
We introduce the measurement problem in quantum mechanics and we briefly \ndiscuss \nthe solutions proposed in literature. We then focus our \nattention on models of spontaneous wavefunction collapse. \nWe describe the two most popular models (GRW, CSL) and list other proposals. \nWe analyze \nin detail a third collapse \nmodel (QMUPL), which is particularly simple (but physically meaningful) to \nbe studied in great mathematical detail. We discuss its main properties. We \nalso describe a "finite \ntemperature" version of this model, which includes dissipative terms. \n \nThese models are Markovian, i.e. the collapse mechanism is driven by a \nwhite noise. Since the ultimate goal is to identify the noise responsible \nfor the \ncollapse with a random field in Nature, it becomes important to study \nnon-Markovian generalizations of collapse models, where the collapsing field \nhas a generic correlation function, likely with a cut off at high \nfrequencies. \n \nModels of this kind have already been studied, as a generalization of the \nCSL model. In this thesis \nwe describe in mathematical detail the generalization of the QMUPL model to \nnon-Markovian noises. After having proved, under suitable conditions, the \nseparation of the center-of-mass and relative \nmotions for a generic ensemble of particles, we focus our analysis on the \ntime evolution of the center of mass \nof an isolated system (free particle case). We compute the explicit \nexpression of the Green's function via the path integral formalism, for a \ngeneric Gaussian noise. We analyze in detail the case of an exponential \ncorrelation function, providing the exact analytical solution. \n \nWe next study the \ntime evolution of average quantities, such as the mean position, momentum \n(which satisfy Ehrefest's theorem) and energy (which is not conserved like \nin the other collapse models). We also compute the non-Markovian master \nequation \nfor an harmonic oscillator, according to this model, and compare \nits structure to the well-known Lindblad structure of Markovian open quantum \nsystems. \n \nWe \neventually specialize to the case of \nGaussian wave functions, and prove that all basic facts about collapse \nmodels (reduction process, amplification mechanism, etc.), which are known \nto be true in the white noise case, hold also in \nthe more general case of non-Markovian dynamics. \n \n \nWe further analyze the evolution \nof Gaussian wave function according to the three different realizations of \nthe QMUPL model so far developed (Markovian, non-Markovian and "finite \ntemperature"), comparing their fundamental features. \nFinally, by analyzing different localization criteria, we set new lower \nbounds on the \nparameters of these models, and we compare them with the upper bounds coming \nfrom known experimental data.
Consider a quantum system prepared in state ψ, a unit vector in a d -dimensional Hilbert space. Let b 1 , ..., b d be an orthonormal basis and suppose that, with some probability 0 < p < 1, ψ 'collapses', i.e., gets replaced by b k (possibly times a phase factor) with Born's probability |〈 b k |ψ〉| 2 . The question we investigate is: how well can any quantum experiment on the system determine afterwards whether a collapse has occurred? The answer depends on how much is known about the initial vector ψ. We provide a number of different results addressing several variants of the question. In each case, no experiment can provide more than rather limited probabilistic information. In case ψ is drawn randomly with uniform distribution over the unit sphere in Hilbert space, no experiment performs better than a blind guess without measurement; that is, no experiment provides any useful information.
Whether wave function collapses or not is a major remaining question in the theory of quantum measurement. This difficulty stems from following two facts. First, it has not been recognized that single-particle quantum mechanics and many-particle quantum mechanics must be treated separately. Second, quantum jump (QJ) and wave function collapse (WFC) need clearer definitions. We define a QJ as a process of selecting a set of system eigenvalues (SEVs) of an observable and a WFC as a process of determining the probability distribution (PD) of SEVs, both from a single measurement. The goal of quantum observation is to obtain the PD, which is determined from an ensemble of SEVs. The wave function becomes an observable when the PD is determined. In single-particle quantum mechanics, a single measurement results in only one set of SEVs and the PD is not observable. Therefore the WFC does not happen. In many-particle quantum mechanics, we focus on the occupation number of a singe quantum state. The wave function does not collapse in general, but there are exceptions. The occupation number can be huge and macroscopic for photons or for Bose-Einstein condensates. In such a case, the PD is determined from a single measurement of a real ensemble and the WFC occurs. We call it a macroscopic quantum jump, which effectively is a measurement of a classical observable.
Whether the wave function collapses or not is a major remaining question in the theory of quantum measurement. This difficulty stems from the following two facts. First, it has not been recognized that single-particle quantum mechanics and many-particle quantum mechanics must be treated separately. Second, quantum jump (QJ) and wave function collapse (WFC) need clearer definitions. We define a QJ as a process of selecting a set of system eigenvalues (SEVs) of an observable and a WFC as a process of determining the probability distribution (PD) of SEVs, both from a single measurement. The goal of quantum observation is to obtain the PD, which is determined from an ensemble of SEVs. The wave function becomes an observable when the PD is determined. In single-particle quantum mechanics, a single measurement results in only one set of SEVs, and the PD is not observable. Therefore, the WFC does not happen. In many-particle quantum mechanics, we focus on the occupation number of a single quantum state. The wave function does not collapse in general, but there are exceptions. The occupation number can be huge and macroscopic for photons or for Bose-Einstein condensates. In such a case, the PD is determined from a single measurement of a real ensemble, and the WFC occurs. We call it a macroscopic quantum jump, which effectively is a measurement of a classical observable.
We are concerned with the problem of detecting with high probability whether a wave function has collapsed or not, in the following framework: A quantum system with a d-dimensional Hilbert space is initially in state ψ; with probability 0 < p < 1, the state collapses relative to the orthonormal basis b1, …, bd. That is, the final state ψ′ is random, it is ψ with probability 1 − p and bk (up to a phase) with p times Born’s probability bk|ψ2. Now an experiment on the system in state ψ′ is desired that provides information about whether or not a collapse has occurred. Elsewhere [C. W. Cowan and R. Tumulka, J. Phys. A: Math. Theor. 47, 195303 (2014)], we identify and discuss the optimal experiment in case that ψ is either known or random with a known probability distribution. Here, we present results about the case that no a priori information about ψ is available, while we regard p and b1, …, bd as known. For certain values of p, we show that the set of ψs for which any experiment E is more reliable than blind guessing is at most half the unit sphere; thus, in this regime, any experiment is of questionable use, if any at all. Remarkably, however, there are other values of p and experiments E such that the set of ψs for which E is more reliable than blind guessing has measure greater than half the sphere, though with a conjectured maximum of 64% of the sphere.
Whether the wave function collapses or not is a major remaining question in the theory of quantum measurement. This difficulty stems from the following two facts. First, it has not been recognized that single-particle quantum mechanics and many-particle quantum mechanics must be treated separately. Second, quantum jump (QJ) and wave function collapse (WFC) need clearer definitions. We define a QJ as a process of selecting a set of system eigenvalues (SEVs) of an observable and a WFC as a process of determining the probability distribution (PD) of SEVs, both from a single measurement. The goal of quantum observation is to obtain the PD, which is determined from an ensemble of SEVs. The wave function becomes an observable when the PD is determined. In single-particle quantum mechanics, a single measurement results in only one set of SEVs, and the PD is not observable. Therefore, the WFC does not happen. In many-particle quantum mechanics, we focus on the occupation number of a single quantum state. The wave function does not collapse in general, but there are exceptions. The occupation number can be huge and macroscopic for photons or for Bose-Einstein condensates. In such a case, the PD is determined from a single measurement of a real ensemble, and the WFC occurs. We call it a macroscopic quantum jump, which effectively is a measurement of a classical observable.