Randomized trial demonstrates bounded-local value recovery failure in Collatz valuation words, implying limitations for digit block recovery.
We study, within the class of Collatz valuation words realized by genuine positive odd integers, whether the exact accelerated digit block q_k at a given position k can be recovered from a bounded local observation: a fixed-width sliding window together with a bounded phase datum. We prove (Unbounded Prefix Memory Theorem) that it cannot: for every window length L, there exist two realized valuation words sharing an identical length-L window and phase at some position k, yet with differing q_k, unconditionally (no bounded-drift hypothesis needed). The construction uses an explicit safe adjacent-symbol swap (odd valuation difference), an exact 2-adic non-integrality argument, and the Backward Lifting construction with an odd-terminal correction to realize both finite truncations by genuine positive odd integers. This is a bounded-local exact-value obstruction, logically distinct from both zero-decision hardness (whether q_k=0 can be decided) and from the fixed-prefix realizability obstruction of the companion Tail-Dependence paper; neither of these stronger claims is established here. A conditional (experimentally supported, not proven) strengthening to explicit on-demand target control, and a conjectural non-ω-regularity of the associated bounded-alphabet ghost language, are also discussed, together with the precise gap remaining to establish them. To the author's knowledge, no prior work establishes a constructive bounded-local separation theorem for the exact accelerated digit block q_k within the class of valuation words realized by positive Collatz integers, of this form; a detailed comparison with the classical parity-vector/2-adic conjugacy literature (Terras, Lagarias, Bernstein, Bernstein-Lagarias), computational-complexity literature (Stérin-Woods), and statistical parity-code literature (Koyuncu et al.) is given in the paper. This paper does not prove the Collatz conjecture. Accompanying code independently re-verifies the paper's explicit 44-symbol / N0 = 89,898,906,153,467,314,169 construction (Section 6) from scratch, and reproduces the qualitative pattern behind the Swap Handle Coverage experimental observation. Part of a four-paper program on Collatz valuation realizability, together with "The Dual Automaton Structure of Collatz Realizability" (DOI: 10.5281/zenodo.21468492), "Tail Dependence and Finite-Address Obstructions in Collatz Valuation Dynamics" (DOI: 10.5281/zenodo.21468378), and "Integer Stabilization and the 2-adic Digit Tower in Collatz Valuation Dynamics" (DOI: 10.5281/zenodo.21468590).
No takes yet. Share an insight, caveat, or question.
KyungUP Moon (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: