Source-linked AI summary
Experimental test of local observer-independence
Massimiliano Proietti, Alexander Pickston, Francesco Graffitti, Peter Barrow, Dmytro Kundys, Cyril Branciard, Martin Ringbauer, Alessandro Fedrizzi
TL;DR
The paper asks whether facts recorded by different observers can be reconciled as observer-independent in quantum mechanics. It tests an extended Wigner’s-friend scenario with six photons and violates the associated inequality by more than five standard deviations, implying observer-dependent interpretation if locality and free choice are maintained.
Problem
The measurement problem leaves unclear how observers’ potentially different records can be reconciled as observer-independent facts.
Source: pp. 1–2Method
The experiment implements an extended Wigner’s-friend scenario in which friends record measurement outcomes inside Alice’s and Bob’s separate laboratories.
Source: pp. 2–3Results
Sexp = 2.416+0.075/−0.075, violating inequality (2) by more than 5 standard deviations.
Source: p. 4Takeaways & Limitations
If locality and free choice hold, the violation means the observers’ facts cannot coexist in a single observer-independent framework.
Source: p. 2The interpretation assumes fair sampling, empirical absence of signalling, and that A0 and B0 measure the friends’ memories as intended.
Source: p. 4
Abstract
from arXiv · showhide
The scientific method relies on facts, established through repeated measurements and agreed upon universally, independently of who observed them. In quantum mechanics, the objectivity of observations is not so clear, most dramatically exposed in Eugene Wigner's eponymous thought experiment where two observers can experience seemingly different realities. The question whether these realities can be reconciled in an observer-independent way has long remained inaccessible to empirical investigation, until recent no-go-theorems constructed an extended Wigner's friend scenario with four observers that allows us to put it to the test. In a state-of-the-art 6-photon experiment, we realise this extended Wigner's friend scenario, experimentally violating the associated Bell-type inequality by 5 standard deviations. If one holds fast to the assumptions of locality and free-choice, this result implies that quantum theory should be interpreted in an observer-dependent way.
SUPPLEMENTARY MATERIALS
The supplementary materials detail the six-photon apparatus and measurement procedures, report a second violation using an alternative A0/B0 protocol, and discuss experimental error sources and loopholes.
Setup details
- Three Sagnac-type sources generate entangled 1550 nm photon pairs; tomography benchmarks the three pairs, and transmission preserves their reported high quality.The pump repetition rate is quadrupled through temporal multiplexing to suppress higher-order emissions.Source: SUPPLEMENTARY MATERIALS · p. 6
Measurement protocol
- Type-I fusion gates perform the friends’ non-destructive photon measurements, encoding each h/v outcome in an ancillary photon that serves as memory.The heralding photon is detected to flag successful fusion-gate operation.Source: SUPPLEMENTARY MATERIALS · p. 6
- A0/B0 are measured by direct projections without a beam splitter, whereas A1/B1 use a beam splitter and Bell-state projections.The 64 settings comprise projections onto eigenstates of Alice’s and Bob’s observables, with six-photon coincidences normalized into relative probabilities.Source: SUPPLEMENTARY MATERIALS · p. 7, Figure S2 · p. 8
Error analysis
- Independent correlator uncertainties are combined, with Monte Carlo and error propagation agreeing within 0.0032; shared normalization in Fig. S3 correlates errors and slightly raises uncertainty.The Monte Carlo routine uses 100,000 samples to account for potentially asymmetric errors at low count rates.Source: SUPPLEMENTARY MATERIALS · p. 8
Error sources
- Higher-order emissions of at least four photon pairs are the main residual error source, and their higher pump-power scaling motivates operation at 100 mW.The experiment’s probabilistic photon-pair sources make multi-pair emissions a primary limit on the achievable Bell-Wigner value.Source: SUPPLEMENTARY MATERIALS · p. 8
Loopholes
- The event-ready design still requires space-like separation of heralding events, setting choices, and the other party’s outcomes to close locality and free-choice loopholes.Closing detection additionally requires photon-number-resolving ancilla detectors and measurements that can project onto all eigenstates in each run.Source: SUPPLEMENTARY MATERIALS · pp. 8–9
Alternative observables
- Sexp = 2.407+0.073/−0.073, again violating inequality (2) by more than 5 standard deviations with the alternative A0/B0 definition.The alternative definition leaves ⟨A1B1⟩ unchanged while changing the other three correlators.Source: SUPPLEMENTARY MATERIALS · p. 9, Figure S3 · p. 10
Alternative measurement protocol
- More than 3 standard deviations of violation remained with the alternative polariser protocol, though about 4.83 ± 0.97% loss reduced the measured correlators.The A0/B0 measurement success probability was reduced by a factor of 1/4, and measurements were normalized to the total counts for ⟨A1B1⟩.Source: SUPPLEMENTARY MATERIALS · pp. 9–10