---
title: Generalized Imaging Theorem
url: https://www.emergentmind.com/topics/generalized-imaging-theorem
type: topic
---

# Generalized Imaging Theorem

Searching arXiv for the specified paper and closely related version to ground the article in the cited literature.
arXiv search query: 1601.02588
The generalized imaging theorem (GIT) is a semiclassical asymptotic result stating that, for a quantum system propagating from a microscopic interaction region to a detector at large distance and time on an atomic scale, the final coordinate-space wave function becomes proportional to the initial momentum-space wave function evaluated at the momentum that is classically mapped to the detector point. In the formulation given by Briggs and Feagin, this establishes that classical-looking trajectories can emerge autonomously from unitary Schrödinger evolution, without requiring environmental decoherence, while still allowing interference whenever the measurement does not isolate a unique trajectory [1601.02588].

## 1. Statement and physical domain

The theorem applies to a quantum system of \(n\) particles propagating from a compact reaction volume at \(t=t_i\) with momentum wave function \(\tilde\Psi(\mathbf p,t_i)\). The propagation may occur under free motion, constant or otherwise deterministic external fields, lensing fields, and mutual particle interactions. The defining asymptotic regime is reached when the accumulated action is large compared with \(\hbar\), and when the propagation time and distance are large compared with microscopic scales, though still within ordinary laboratory settings [1601.02588].

Its central claim is that the asymptotic spatial wave function of any multi-particle quantum system becomes proportional to the initial momentum wave function, with position and momentum coordinates connected by the classical equations of motion. In the one-sentence summary given in the related earlier presentation, after sufficiently long deterministic propagation, a quantum particle’s final spatial wave function is essentially the initial momentum wave function evaluated at the momentum corresponding to the classical trajectory connecting source and detector, so measured positions and momenta obey classical mechanics even without decoherence [1506.04287].

This scope is broader than the older imaging theorem of scattering theory because the GIT is formulated for arbitrary nonrelativistic multi-particle motion in external deterministic fields and with particle interactions. In the many-body case, the mapping is not merely from one particle’s source momentum to one detector coordinate; it is a mapping of the entire many-body phase-space configuration via a classical trajectory bundle [1601.02588].

## 2. Mathematical form of the theorem

The theorem is expressed by the asymptotic relation
\[
\Psi(\mathbf r_f,t_f)\approx (2\pi\hbar)^{3n/2}\,K_{sc}(\mathbf r_f,t_f;\mathbf r_i,t_i)\,\tilde\Psi(\mathbf p_i,t_i),
\]
where \(\mathbf r_i,\mathbf p_i\), and \(\mathbf r_f\) are related by the classical equations of motion. The corresponding probability relation is
\[
|\Psi(\mathbf r_f,t_f)|^2 \, d\mathbf r_f \approx |\tilde\Psi(\mathbf p_i,t_i)|^2 \, d\mathbf p_i,
\]
or equivalently,
\[
|\Psi(\mathbf r_f,t_f)|^2 \approx \frac{d\mathbf p_i}{d\mathbf r_f}\, |\tilde\Psi(\mathbf p_i,t_i)|^2.
\]
This is the key result: quantum propagation over sufficiently large distances and times turns the detection probability into a classical trajectory-density mapping of the initial momentum distribution [1601.02588].

The classical factor in this mapping is the Van Vleck determinant, which acts as the classical trajectory density. In the coordinate-space semiclassical propagator,
\[
K_{sc}(\mathbf r_f,t_f;\mathbf r_i,t_i) =(2\pi i\hbar)^{-3n/2} \left|\det \frac{\partial^2 S_c}{\partial \mathbf r_f \partial \mathbf r_i}\right|^{1/2} \exp\!\left(\frac{i}{\hbar}S_c(\mathbf r_f,t_f;\mathbf r_i,t_i)\right),
\]
the determinant satisfies
\[
\left|\det \frac{\partial^2 S_c}{\partial \mathbf r_f \partial \mathbf r_i}\right| \sim \frac{d\mathbf p_i}{d\mathbf r_f}.
\]
Accordingly, the wave function at the detector is the initial momentum amplitude transported along a classical path and weighted by the classical focusing or defocusing factor [1601.02588].

A useful conceptual distinction follows directly from this formula. Ehrenfest’s theorem concerns the classical motion of expectation values, whereas the GIT concerns individual detection probabilities along classical paths. This is why the papers describe the relation \(x_f \leftrightarrow p_i\) as deterministic in the asymptotic regime: each detected position corresponds to a definite classical trajectory and to a specific initial momentum on that trajectory [1506.04287].

## 3. Derivation from semiclassical propagation

The derivation starts from mixed coordinate-momentum propagation,
\[
\Psi(\mathbf r_f,t_f)=\int d\mathbf p\, \tilde K(\mathbf r_f,t_f;\mathbf p,t_i)\,\tilde\Psi(\mathbf p,t_i),
\]
with
\[
\tilde K(\mathbf r_f,t_f;\mathbf p,t_i)=\langle \mathbf r_f|U(t_f,t_i)|\mathbf p\rangle.
\]
For large action, the mixed propagator is approximated semiclassically as
\[
\tilde K_{sc}(\mathbf r_f,t_f;\mathbf p,t_i) =(2\pi i\hbar)^{-3n/2} \left|\det \frac{\partial^2 \tilde S_c}{\partial \mathbf r_f \partial \mathbf p}\right|^{1/2} \exp\!\left(\frac{i}{\hbar}\tilde S_c(\mathbf r_f,t_f;\mathbf p,t_i)\right),
\]
where the mixed action \(\tilde S_c\) is related to the coordinate-space classical action by a Legendre transformation [1601.02588].

The stationary-phase condition selects the classical initial momentum:
\[
\frac{\partial \tilde S_c}{\partial \mathbf p}= \mathbf r - \mathbf r_i =0.
\]
This is the point at which the classical trajectory enters the quantum calculation. Evaluating the momentum integral by stationary phase yields the asymptotic form of the theorem, with \(\mathbf p_i\) fixed by the classical trajectory connecting \(\mathbf r_i\) and \(\mathbf r_f\) [1601.02588].

The earlier presentation states the same logic in one-dimensional notation. There the mixed action \(S_c(x_f,t_f;p,t_i)\) is related to the coordinate-space action \(S_c(x_f,t_f;x_i,t_i)\) by
\[
S_c(x_f,t_f;p,t_i) = S_c(x_f,t_f;x_i,t_i) + p\cdot(x-x_i),
\]
and stationary phase imposes
\[
\frac{\partial S_c}{\partial p} = x - x_i = 0.
\]
This gives the same outcome: the asymptotic coordinate-space wave function is proportional to the initial momentum-space wave function evaluated at the classically selected momentum [1506.04287].

## 4. Why the theorem is “generalized”

The theorem is termed “generalized” because it extends the older imaging theorem from simpler scattering and free-motion settings to multi-particle systems in \(3n\)-dimensional phase space, arbitrary deterministic external fields, particle-particle interactions, and more general laboratory detection geometries [1601.02588].

In that generalized setting, the theorem does not merely describe a far-field approximation for a single particle. It provides a classical image of the initial momentum distribution in a higher-dimensional many-body space. The final detection pattern is therefore a classical image of the source momentum distribution, but one encoded within the semiclassical structure of the quantum propagator rather than derived from a collapse or localization postulate [1601.02588].

The physical assumptions are correspondingly explicit. The particles originate in a microscopic source region, propagate for distances and times large on an atomic scale, and evolve under deterministic Hamiltonian dynamics. No environment is needed in the derivation: the result follows from unitary Schrödinger evolution alone. If environmental interaction is present, decoherence may act on top of this mechanism, but the GIT itself does not require a decohering bath or stochastic environment [1506.04287].

This suggests a specific sense in which the theorem describes an autonomous quantum-to-classical transition. The emergence of trajectory-based detection does not require abandoning wave mechanics; rather, classical mechanics appears through the asymptotic semiclassical transport law already contained in the propagator [1601.02588].

## 5. Canonical examples and onset of validity

The simplest example is the free particle in one dimension. For a Gaussian wave packet, the exact wave function approaches the imaging-theorem form at large \(t\):
\[
\Psi(z_f,t_f)\approx \left(\frac{m}{it}\right)^{1/2} \exp\!\left[i\frac{m z_f^2}{2\hbar t}\right]\tilde\Psi(p_i), \qquad p_i=\frac{m z_f}{t}.
\]
This exhibits the central relation directly: the position-space wave function at time \(t\) is proportional to the initial momentum wave function evaluated at the momentum that would classically bring the particle to \(z_f\) [1601.02588].

The earlier version gives the same free-particle result in the form
\[
x_f = \frac{pt}{m}, \qquad \frac{dp_i}{dx_f} = \frac{m}{t},
\]
and therefore
\[
\Psi(x_f,t) = \left(\frac{m}{it}\right)^{1/2} \exp\!\left(\frac{i m x_f^2}{2\hbar t}\right) \widetilde{\Psi}\!\left(\frac{mx_f}{t},0\right).
\]
For an initial Gaussian packet of width \(\sigma\), the exact free evolution reaches the asymptotic imaging-theorem regime when
\[
\frac{\hbar t}{m} \gg \sigma^2.
\]
The condition is equivalently presented as
\[
\left(\frac{\hbar t}{m}\right)^{1/2} \gg \sigma,
\]
which characterizes the onset of the transition zone [1506.04287].

A second explicit example is propagation under a constant force \(F\), for which the classical mapping becomes
\[
p_i = m\left[\frac{z_f-z_i}{t}-\frac{F t}{2m}\right], \qquad z_f = z_i + \frac{p_i t}{m} + \frac{F t^2}{2m}.
\]
The asymptotic wave function again reduces to the initial momentum wave function, now with an additional field-dependent phase. The Van Vleck factor remains \(m/t\), so the probability transport remains classically governed [1601.02588].

The onset of validity can occur at surprisingly small microscopic scales. For realistic atomic processes such as dissociation of \(H_2\) and ionization of hydrogen, the transition to imaging-theorem behavior may begin already around distances of order \(10^2\) atomic units, with correspondingly short times. The papers emphasize that this is still microscopic, even though the propagation is already asymptotic for semiclassical purposes [1601.02588].

## 6. Measurement, interference, and the relation to decoherence

The theorem has direct significance for coincidence and time-of-flight measurements. If one measures both final position and momentum for each fragment, or reconstructs them from time-of-flight and detector position, then the classical mapping is complete and each detected event corresponds to a unique classical trajectory back to the reaction zone. This provides a theoretical basis for the widespread experimental practice of tracing detected particles backward with classical mechanics [1601.02588].

The theorem does not imply that interference disappears. If the measurement is incomplete—if only position is measured, or if not enough information is available to isolate a unique classical path—then more than one classical trajectory can contribute to the same detection outcome. The observed amplitude is then a coherent sum over path contributions, and interference can occur. The paper illustrates this with atom interferometry:
\[
\Psi(\mathbf r_f,t_f)\propto \left(\frac{m}{it}\right)^{3/2} \left[ e^{iS_0^{(0)}/\hbar}\tilde\Psi(\mathbf p_0) + e^{iS_0^{(1)}/\hbar}\tilde\Psi(\mathbf p_0+\mathbf p_g) \right],
\]
leading to fringes
\[
|\Psi(\mathbf r_f,t_f)|^2 \propto \left(\frac{p_0}{2L}\right)^3 2N^2 \left[1+\cos\left(\frac{2\pi}{d}x\right)\right].
\]
Thus the GIT is fully compatible with interference: classical trajectory mapping governs propagation, while quantum superposition survives whenever multiple mapped paths remain unresolved [1601.02588].

This point underlies the theorem’s contrast with decoherence-based accounts. In the decoherence view, classicality arises because environmental interactions suppress off-diagonal density-matrix elements. In the GIT view, trajectory structure appears already from unitary Schrödinger evolution in the asymptotic semiclassical regime, even in perfect vacuum. The papers do not deny decoherence; rather, they argue that it is not the sole or primary reason that detected motion looks classical. The diagonal density-matrix elements become classical because of the imaging theorem, while off-diagonal terms may oscillate and average out with finite resolution [1601.02588].

A common misconception is therefore that classical-looking detector patterns necessarily demonstrate environment-induced decoherence. The theorem states a narrower and more technical claim: classical detector patterns can emerge spontaneously from deterministic unitary propagation, whereas interference remains available whenever the measurement resolves coherent superpositions rather than a unique classical trajectory [1506.04287].

Source: https://www.emergentmind.com/topics/generalized-imaging-theorem