Answers are grounded in this paper and cite their sources.
Steretic logic formalises informative absence: a missing link between two nodes is structurally required when they share at least two common neighbours. We apply this requirement, unchanged, to the link graph of causal sets and ask what the resulting "steretic voids" tell us about spacetime curvature. Void counts respond to curvature in every geometry we tried: two- and four-dimensional spaces of constant curvature, a static conformally flat four-dimensional spacetime, and the non-conformally-flat product R x S^2 x S^1, in which curvature changes the causal order itself. In the four-dimensional local tests the response has a uniform sign: counts fall where the Ricci scalar is positive. The response is not local, however. (i) In flat space every local count is an exact function of rho^(1/4) times the proper distances to the boundaries of the region; for near spacelike voids the leading term is proportional to the product of the past and future distances. (ii) In the conformally flat test a one-parameter model of density averaged along light cones predicts the response at a second wavenumber better than a local law, with an averaging length comparable to the size of the region rather than to the discreteness scale. (iii) In the R x S^2 x S^1 test the response grows with the thickness of the region as R lambda^2, where lambda is the distance to the boundaries; a purely local model is rejected (links: chi^2 = 353 for 35 degrees of freedom against 22 for 33 with the lambda^2 terms; near voids: 46.5 against 15.2). Plain link counts carry the same curvature information with better resolution, and the smeared Benincasa-Dowker estimator is not converged in the shallow regions accessible to us. In particular, near spacelike voids are not a local curvature probe at these sizes. What remains is a set of scaling laws, a null-extended picture of the response, the conjecture that such counts measure curvature integrated along light rays, and the methodological lesson that locality must be tested by varying the size of the region.
No takes yet. Share an insight, caveat, or question.
Oleksandr Krol (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: