26th Australian Institute of Physics Congress and the Australian and New Zealand Conference on Optics and Photonics

Neglected Distinguishability is Not Indistinguishability: Hidden Entanglement in Post-Selected Interference

Not scheduled

20m

Belinda Hutchinson Building (The University of Sydney )

Contributed Oral AIP | Quantum Science and Technology (QST)

Speaker

Tony Yan (The Australian National University)

Description

Interference between multiple spontaneous parametric down-conversion sources produces phase-dependent two-photon coincidence fringes with no single-source analogue. In frustrated-interference and path-identity experiments, a coincidence channel can go dark through destructive interference between source amplitudes. In recent three- and four-source versions, Jiang et al., Nat. Commun. 2026 claim "subjective" which-path information, while Wang et al., Sci. Adv. 2025 report a Bell violation with "unentangled photons".

We present a source-tracked mode-overlap model. The literature analysis reproduces the fringe by treating photons from different crystals as a single detected mode, hiding the source-origin labels carried by the emitted fields. In our description, the mode overlap sets how much source-labelled field enters the detected mode, while detection is a projection onto the detectable subspace. The orthogonal sector is discarded in the literature, along with the which-source information and entanglement it carries.

With the labels kept, the observed fringes are still recovered. A dark fringe is therefore a statement about the postselected modes, not by itself evidence that the sources did not emit. Where frustrated interference reads the dark fringe as cancelled emission, the matched amplitude can be coherently reabsorbed rather than interpreted as a source being "switched off". Modes treated as loss can retain which-source information, and detectors sampling those modes would reveal the discarded entanglement - a quantum-eraser picture.

Applying this, we recast Jiang et al.'s three-source grouping "contradiction" as ordinary interference: the two groupings are not independent alternatives, but partitions of one coherent probability amplitude. We show that Wang et al.'s detected product state is only the postselected sector; source-origin entanglement is removed by postselection before the claimed Bell analysis. We also propose a spatially resolved coincidence test: scanning the beam alignment tunes the mode overlap and maps the trade-off between fringe visibility and which-origin distinguishability.