Answers are grounded in this paper and cite their sources.
Theoretical analysis reveals span growth behavior shaped by capacity decay in Heisenberg graphs, suggesting new insights into geometric dynamics.
The spectral admissibility programme measures a projective block capacity σc(n) along the breadth-first cascade of the Heisenberg Cayley graph, decaying as a power law σc(n) ~ a\,n^-δc, and converts capacity exponents into cascade growth exponents through the structural relation β = 1/(δ + 1/2), derived on Ramanujan–Lubotzky–Phillips–Sarnak (LPS) relaxation graphs. The present note derives the span-growth law that the Heisenberg substrate itself imposes. The cumulative fingerprint span $r(n)$ obeys the exact increment identity Δ r(n) = σc(n)\,|Sₙ|, and the exact breadth-first spheres of Heis₃(Z) grow polynomially with homogeneous dimension $D = 4$. Asymptotic summation then yields a complete three-branch classification, depending only on (δc, D): polynomial span growth r(n) ~ n^D-δc for δc < D; logarithmic growth for δc = D; saturation r_∞ - r(n) ~ n^-(δc - D) for δc > D. Two structural consequences follow. First, the span is never proportional to the number of configurations reached: r(n)/|Bₙ| → 0 for every δc > 0, so the valence–exploration proportionality that closes the LPS growth equation does not transpose to this substrate. Second, no uniform constant c can sustain the LPS-form bound Δ r ≤ c\,σc √r: since r(n) ≤ |Bₙ|, any such constant must exceed |Sₙ|/√|Bₙ| nD/2-1, which diverges for every polynomial growth dimension $D > 2$. The square-root frontier factor — and with it the additive 1/2 of the cascade-exponent law — is therefore specific to expander-type substrates; on Heisenberg graphs the capacity-to-growth map on the growth branch is β = D - δc. Pipeline data illustrate the classification: the window-effective exponents make contact near $q = 211$ (δcᵉᶠᶠ = 3.748 vs Deff = 3.753) and separate strictly for q ≥ 307, with δcᵉᶠᶠ - Deff between $-0.21$ and $-0.49$ at $6.7$–$14.5$ standard errors of the inter-pair bootstrap, a sign stable across all $27$ window-shift and knee-trimming variants and under both logarithmic conventions. The contact is therefore transitional in the accessible data — the span sits in the growth branch at the largest measured primes — and no current indication forces δc = D; the statement concerns window-effective exponents only, with no asymptotic claim. The polynomial span exponent studied here is distinct from the exponential per-shell rate carried by the symbol β* in the projected-Yukawa mass factorisation; the two objects are separated explicitly. Interpretive status. The classification theorem and the two no-go corollaries are proved under explicit power-law hypotheses; the numerical exponents are finite-window measurements; the $q = 211$ near-contact is resolved, within the accessible data, as a transitional crossing of window-effective exponents, while the asymptotic value of δc remains open. As a structural reading, span growth on an emergent relational substrate is governed by a competition between geometric opening (volume growth) and capacity decay (novelty exhaustion); at the measured windows the two pass through near-balance transitionally and separate at the largest measured primes. This reading is an interpretation, not a result.
No takes yet. Share an insight, caveat, or question.
Jérôme Beau (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: