Antonio Di Lorenzo
5 min
The study investigates the concept of the quantum Cheshire cat, a phenomenon where a particle's physical properties (like polarization, the 'grin') appear to be spatially separated from the particle itself (the 'cat'). Previous proposals suggested that detecting this disembodiment simply requires observing local averages of the cat and grin observables to be unity. The author challenges this, arguing that local averages are insufficient and can lead to misleading conclusions. Instead, the paper derives the exact probability distribution and characteristic function for arbitrary coupling strengths, using a cross-average of the two observables as a more reliable indicator of the phenomenon.
The author shows that the quantum Cheshire cat is fundamentally a consequence of quantum interference. By analyzing the cross-average of the 'cat' and 'grin' measurements, the paper demonstrates that this disembodiment is a common occurrence in post-selected measurements. Crucially, the study finds that the phenomenon is not limited to weak measurements but can also occur at intermediate coupling strengths. The paper provides an operational definition for the Cheshire cat parameter, which can be calculated from experimental data by considering both successful and unsuccessful post-selection events.
This work clarifies the theoretical underpinnings of the quantum Cheshire cat, moving beyond simple local averages that may be misinterpreted. By providing a more rigorous, interference-based criterion, the paper offers researchers a clearer way to experimentally verify the separation of physical properties from particles. This is significant for understanding the limits of weak measurement and the role of post-selection in quantum mechanics.
We analyze the proposal of Aharonov, Popescu, Rohrlich and Skrzypczyk [New. J. Phys. 15, 113015] of disembodying physical properties from particles. We argue that a different criterion, based on the cross-average $\langle \mathrm{`}cat\ somewhere \mbox{'}\times \mathrm{`}grin\ somewhere\ else\mbox{'}\rangle$ should be used to detect the disembodiment, rather than the local averages $\langle \mathrm{`}cat\ somewhere \mbox{'}\rangle$ and $\langle\mathrm{`}grin\ somewhere\ else\mbox{'}\rangle$. Here, the exact probability distribution and its characteristic function are derived for arbitrary coupling strength, preparation and post-selection. This allows to successfully hunt down the quantum Cheshire cat, showing that it is a consequence of interference, that it is present also for intermediate-strength measurements, and that it is a rather common occurrence in post-selected measurements.
Sam: There's also something in the paper about how you handle the failed post-selection events—the runs where you don't catch the "cat." How does that factor in?
Alex: This is one of the more counterintuitive moves in the paper. The standard approach discards those events—you post-select on successful outcomes and analyze those. Di Lorenzo argues you should instead assign a negative weight to the unsuccessful post-selections and include them. When you do that, the interference terms balance properly rather than vanishing, and the Cheshire Cat parameter becomes statistically robust across the full ensemble rather than just the selected subset.
Sam: You're using the cases where the experiment "failed" to stabilize the result.
Alex: It's less about failure and more about completeness. Post-selection always throws away information. By accounting for what you discarded—even with a negative sign—you recover a more complete picture of the interference structure. The math works out cleanly; it's the interpretation that takes some adjustment.
Sam: What are the limits of this framework? Where does it leave things unresolved?
Alex: The honest answer is that the framework still rests on subjective state reduction. The time asymmetry introduced by the observer's post-selection is a postulate, not something derived from first principles. That's a known tension in quantum foundations generally, but it means the operational prescription Di Lorenzo provides—however clean—doesn't resolve the deeper question of what the observer's role actually is. The paper moves the conversation from "did we observe disembodiment" to "here is how to correctly quantify it," but it doesn't close the loop on what state retrodiction means physically.
Sam: So the contribution is clarifying the measurement protocol and identifying the right observable, while the foundational question about the observer remains open.
Alex: That's a fair summary. And it points toward where the work goes next—applying this cross-correlation framework to multi-particle systems or higher-dimensional Hilbert spaces, where the interference structure is richer and the distinction between classical co-occurrence and genuine quantum correlation becomes even harder to establish by other means.
Sam: It's a meaningful shift in how the field approaches these experiments. Less "did we see the cat's grin" and more "here's the observable that actually tells you whether the grin is real."
Alex: Precisely. And getting that observable right is what makes the difference between a reproducible result and an artifact of the measurement design. Thanks for listening to ResearchPod.