Frame-Budget-Ansatz (FBA)
Was ist Zeit — operativ gedacht?
Frame-Budget Approach (FBA)
What is time — operationally?

Bell Violation Without Entanglement? FBA Handles for Postselection & Channel Tests

•

, , ,

•

5 min read

In https://pmc.ncbi.nlm.nih.gov/articles/PMC12315985/ (Wang et al., Science Advances, 2025), it is claimed that a Bell inequality is violated in a four-photon interference setup, even though the photons involved are not described as entangled within the system. The report names quantum indistinguishability via path identity as the mechanism and uses postselection on four-fold coincidences; the Bell parameter is given as S: 2.275 ± 0.057.

The FBA view translates this not into “wilder quantum magic,” but into protocol questions: Which steps are the channel, which are conditioning/selection filters, and which pass/fail null tests decide whether this is truly a Bell situation—or merely a selectively normalized subensemble being tested.

Categories

  • Contribution type: Experiment
  • Topics: C4 (Quantum information & channels), C8 (Methodology, data & reproducibility), C3 (Spacetime, locality & QFT)

Source anchors & subject

Submitted link

https://flip.it/9AoGgP

Primary sources

Reality check

  • Standard/established: The full text reports a CHSH evaluation from four-fold coincidences and gives a Bell parameter S: 2.275 ± 0.057 (interpreted as more than four standard deviations above the classical bound).
  • Standard/established: The setup uses multiphoton “frustrated interference” with path identity; the relevant signature arises in the postselected four-photon term, and the local two-fold count rates are described as showing no interference.
  • Hypothesis: Whether this may be read as a “Bell test” in the strict operational sense (sample definition, normalization, selection-independence, timing/loopholes) is disputed; counter-analyses attribute the apparent violation to postselection and normalization.

FBA view

  • Handle: “Unentangled” is only meaningful once it is clear which operational notion of preparation is meant; postselected subensembles are a different operational state than the overall preparation. (Definition III.3.1.1)
  • Principle: A Bell evaluation is only as good as the underlying measurement description; FBA asks: does S come from well-defined outcome probabilities (POVM statistics), or from rates that are normalized after the fact? (Definition III.3.2.1)
  • Proxy: Model the full setup as a composition of admissible maps plus explicit conditioning; once a selection step is not trace-preserving, “Bell-like” correlations can appear as protocol artifacts. (Definition III.4.1.1)
  • Confounder: Postselection on four-fold coincidences can create collider bias: the selection can (covertly) couple settings and internal variables even if the optics are local.
  • Control idea: No-signalling must be tested on the same sample definition on which S is computed; “local counts without interference” is an indication, but not a substitute for a marginal test under the same conditioning. (Lemma III.5.2.1)
  • Residual: Reproducibility here means: the analysis must be invariant under refinements (windows, cuts, normalization), or it must disclose the dependence; otherwise the conclusion is protocol-fragile. (Definition I.8.3.2)

New insights from FBA

  • FROM→TO: “Bell violation without entanglement” → “Bell-like correlation in a selected subensemble.” Implicit assumption: the normalization corresponds to a genuine measurement probability and the selection is settings-independent.
  • FROM→TO: “Product state” → “product state on the full register, but not on the conditioned register.” Implicit assumption: conditioning does not change resource attribution (what counts as the “state”).
  • FROM→TO: “Local rates are flat” → “no-signalling is satisfied.” Implicit assumption: marginals are tested on exactly the same event set as S (same window/selection/normalization).
  • FROM→TO: “Robust across runs” → “robust against window/cut choice.” Implicit assumption: the result is refinement-invariant, or the dependence is part of the claim. (Definition I.8.3.2)

Alternative readings & conclusions

  • Standard/established: Multiphoton interference via path identity is a real, nonclassical correlation phenomenon and is relevant as a platform for quantum optics/metrology.
  • Hypothesis: The reported CHSH violation arises primarily from postselection and (Bell-atypical) normalization; as a statement about local realism this would then be a protocol confusion—not “nonlocality without entanglement.”
  • open/unclear: Which Bell assumptions are operationally satisfied (timing of settings relative to selection, event readiness, sample bias, random choice, independent registers) cannot be decided unambiguously without a fully disclosed protocol register.

Tests/Experiments (Pass/Fail) with an FBA touch

  • Null test (FBA): S under permutation of the settings labels | Dryad raw data + recompute | S drops toward the classical bound | S remains similarly high despite label permutation
  • Residual (Hypothesis): S as a function of coincidence window | recompute with varying time windows | S is stable across a reasonable window width | S shows a narrow “tuning” peak or strong window dependence
  • Null test (Hypothesis): marginal drift per party on the same sample | compute local marginals from raw counts under the same selection | marginals remain independent of the other party’s settings | systematic marginal shift with the other party’s settings
  • Pass/Fail (open/unclear): timing protocol for settings vs selection | experiment logs/time-tags | settings choice is strictly separated from the selection criterion | selection effectively influences the settings distribution (or vice versa)

Added value of the FBA view

Added value: 8/10 – The FBA framework cleanly separates “interesting interference physics” from the “Bell claim” and provides a short, data-driven null-test suite that makes postselection/normalization visible as operational control knobs.

Reference list (URL-only)

Leave a Reply

Your email address will not be published. Required fields are marked *