Feynman published his path-integral formulation in 1948, describing how a quantum system evolves from an initial to a final state by summing over every possible trajectory. The approach became foundational for quantum field theory, cosmology and condensed-matter physics — but its two underlying postulates had never been directly tested in the laboratory.
A team at South China Normal University in Guangzhou has now done that. Writing in Science Advances, the researchers report measuring the probability amplitudes of more than 1.4 million possible paths taken by single photons in an optical system. They divided the region between a start point A and an end point B into a grid, connected adjacent points with line segments, and measured the propagator for each segment.
The first postulate — that a quantum particle does not follow a single trajectory, but that every probable path contributes — was validated with a mean absolute percentage error of 4.45% and a fidelity of 94.9%. The second — that all paths carry equal-magnitude amplitudes differing only by a phase given by the classical action — was confirmed with 94.7% fidelity.
"Because there were so many possible paths that the photons could take, the primary challenge in our experiments was to achieve sufficient accuracy to reconstruct millions of path amplitudes, since even marginal errors in individual propagator measurements accumulate multiplicatively, destroying phase coherence and rendering the reconstructed path distribution nearly random," said first author Yong-Li Wen. The team raised single-photon propagator fidelity from 87.6% to 98.5% using signal amplification, a custom high-precision imaging system, real-time normalisation with a reference beam, and nanoscale mechanical stability.
Group leader Shi-Liang Zhu says the result supports the view that quantum paths "reflect physical reality rather than being mere mathematical artefacts". The propagator technique could be extended to study quantum-to-classical transitions, decoherence mechanisms, entangled histories and indefinite causal order across multiple space-time points — and, further out, to precise propagator measurements in interacting systems and curved space-times.
