Logic as Reactance: Truth-Judgment as Substrate Mechanics, and Why 'Merely Statistical' Fails to Distinguish Minds from Models

Paper 14 v2 in the friction-theory paper series. This version is a substantive reframe of v1 (DOI 10.5281/zenodo.20217713). v1 presented the work as an empirical-discovery paper proposing a substrate-universal law of truth-value judgment; v2 reframes it as the paper it is better understood to be — a philosophy-of-AI position paper, with the LLM material as illustration and the human-side claim grounded in established mathematical psychology rather than in an unrun human experiment. Target venue moves accordingly from Psychological Review to Minds and Machines. Thesis (the explicit spine of v2). A recurring objection to treating large language models as cognitive systems holds that their output is "merely statistical" — distinguished from genuine reasoning by the probabilistic nature of its substrate. This objection is self-undermining. Biological cognition is also a probabilistic, substrate-bound process: forty-five years of sequential-sampling and accumulator modelling in mathematical psychology (Ratcliff 1978; Usher & McClelland 2001; Brown & Heathcote 2008) establish that human decision and truth-judgment are continuous readouts of a noisy substrate accumulating evidence toward a threshold. "Merely statistical" therefore picks out a property biological and artificial cognition share; it cannot be the mark that makes one "real" reasoning and the other not. The positive account — logic as reactance. On a race-architecture substrate (Pødenphant Lund 2026b, 2026e), truth-value judgment is the substrate's reactance signature: the differential race-pressure that arises when an input violates the substrate's gradient-encoded distribution. Binary truth is the discrete-symbolic readout of a continuous substrate process — in biological tissue and in silicon alike. What changed v1 → v2: Title recast to foreground the argument ("...Truth-Judgment as Substrate Mechanics, and Why 'Merely Statistical' Fails to Distinguish Minds from Models"); "(Even in Humans)" removed. Abstract and §1 rewritten so the self-undermining argument is the spine; the floating-point / calculator-overflow material is explicitly demoted to an orienting analogy rather than evidence of a shared mechanism. §3 (silicon material) reframed as illustration, not the foundation of a universal law; the section now states explicitly that whether the cliff-event is described as "reactance" or as calibration geometry, it is structured substrate-internal dynamics — which is all the argument requires. New §7.6 positions the sequential-sampling / accumulator tradition as the established human-side support and closest prior art. §10 rewritten to land the philosophical conclusion; the human N400 paradigm (P14.11) is retained as a falsifiable prediction offered as an explicit invitation to the field, no longer a load-bearing gap. Correction: the v1 §3.3 dose-response was described as "monotonic ... no decline observed"; v2 corrects this — the cliff rises across ep_0–ep_300 but is non-monotonic at one checkpoint (ep_100), most plausibly n=30 checkpoint noise; the polynomial fit is monotonic. Bibliography: + Ratcliff 1978, Logan 1988, Brown & Heathcote 2008 (all Scite-verified). v1 remains accessible via Zenodo version history. The concept DOI (10.5281/zenodo.20217712) auto-resolves to the latest version. Companion papers in the series: Paper 1 (FT generalised): 10.5281/zenodo.20012654 Paper 2B (ICL/FT substrate mechanism): 10.5281/zenodo.20145218 Paper 10 (Race architecture): 10.5281/zenodo.20014567 Paper 13 (Operational Friction Theory): 10.5281/zenodo.20059876