Nobody is born with a diagnosis; you are born with a biological starting point, and whether it tips into a diagnosis depends partly on small, measurable shortfalls that can sometimes be corrected. This paper is about the large population that carries a vulnerable substrate-configuration without ever crossing a diagnostic threshold, and about supporting that configuration where the rate-limiting step is cofactor-availability, on the pre-diagnostic side of the prevention-cure axis. Abstract. Clinical diagnoses are outcome-clusters: shared symptom-pictures sufficiently similar to receive a common label. The substrate that produces them — the configuration of genes, cofactor-status, and enzyme-capacity that determines how an individual's race-resolution architecture computes — is the more fundamental level, and is not identical with the diagnosis. This paper takes that distinction to its logical conclusion: if the substrate is fundamental and the diagnosis is only the outcome-cluster, there must exist a large population of individuals who carry a vulnerable substrate-configuration without crossing any diagnostic threshold — sub-threshold individuals, invisible to diagnosis-first research because that research recruits on diagnostic criteria. Their substrate-vulnerabilities are nonetheless real and, where rate-limiting sits at cofactor-availability, cofactor-supportable. Three-claim decomposition. The paper makes three claims with different evidence-bases, distinguished explicitly: (1) empirically well-supported: clinical diagnoses are outcome-clusters; the substrate is the more fundamental level; a large sub-threshold population exists (anchored on Zhang 2023 meta-analysis n = 1,129,969; Linscott Merikangas 2007; Sasson 2013; Ruzich 2015); (2) mechanistically plausible: sub-threshold substrate-vulnerabilities are real and, where rate-limiting sits at cofactor-availability, supportable by cofactor-intervention matched to measured substrate-profile (anchored on Gilbody 2007 MTHFR HuGE review; Maruf 2022 L-methylfolate meta-analysis; Tiemeier 2002 Rotterdam); (3) the empirical question the paper invites tests for: cofactor-support matched to measured substrate-profile produces the predicted preventive effect at clinically-meaningful effect-size. P1 and P5 adjudicate claim 2; P6 adjudicates claim 3. Nine falsifiable predictions. P1 (pivotal): carrier × cofactor-status interaction predicts sub-threshold versus manifest. P5 (pivotal + framework-distinct): cofactor-support is null without bottleneck; forward-looking stratified re-analysis of MoodFOOD (Bot 2019 JAMA) by MTHFR-genotype + baseline folate / B12 / homocysteine status predicted to recover the sub-population effect that the unstratified analysis dilutes toward null. P7 (framework-distinct): Category B hysteretic-consolidation durability profile after planned withdrawal. P8 (framework-distinct): hysteresis-accumulation on compensating routes during untreated sub-threshold years produces displacement-comorbidity whose route-identity depends on environmental reinforcement-availability (e.g., ADHD sub-threshold with alcohol access → SUD; without → over-control / burnout). P9 (new, dialog-driven 2026-05-28): behavioural-cluster classification carries lower-information but non-zero probabilistic evidence about substrate-vector; combination of behavioural + direct-measurement predicted to outperform either alone. Discipline-point: measurement, not inference, with explicit behavioural-evidence nuancing (§7.1 and §7.1b). The substrate-profile is measured (blood test, gene test). The paper does not license symptom-to-supplement inference. The behavioural-evidence nuancing of §7.1b acknowledges that existing diagnostic methodology (DSM, dimensional self-report, structured interviews) operates in a lower-information regime whose limits are framework-explicable: behavioural classification carries population-level cluster-allocation value and probabilistic individual-level information about substrate-vector, but direct measurement remains framework-preferred for individual-level intervention-matching decisions where route-specific resolution is the operational question. Author position and invitation. The author is an independent researcher without clinical credentials and without institutional laboratory access. The framework is positioned as hypothesis-and-invitation: it specifies what should be tested, developing the one-carbon / folate–B12–MTHFR cycle as the worked exemplar (strongest gene × cofactor evidence) with the other cofactor-axes catalogued as candidate generalisations rather than individually defended. Several predictions are testable on existing public-access cohort data without new collection: the cofactor-stratified pivotal tests (P1, P5) require a cohort that measured one-carbon status — the Rotterdam Study is the primary anchor, since large genotyped biobanks such as UK Biobank did not assay folate, B12, or homocysteine and so support the genetic-main-effect and behavioural-versus-substrate components (P9) rather than the cofactor stratification — with Whitehall II, MIDUS, ABCD, HUNT, and the Swedish national-register cohort supporting the remaining predictions. We invite collaboration from behavioural geneticists, nutritional-psychiatry researchers, preventive-medicine programmes, biobank-based epidemiology groups, and family-cohort researchers. The framework is a research-programme; testing it cannot be completed within any single tradition. We make no patient-level recommendations. Companion papers in the Friction Theory series: Paper 8 (Psychiatric Clinical Intervention): 10.5281/zenodo.20059866 (reserved) Paper 8b (Compound Race Pathology — framework): 10.5281/zenodo.20059870 (reserved) Paper 8c (Translational Research-Programme): 10.5281/zenodo.20059872 (reserved) Paper 0 (Behavioural Friction Theory): 10.5281/zenodo.19462499 Paper 1 (Friction Theory substrate): 10.5281/zenodo.20012654 Series position. Paper 8d in the Friction Theory paper-series. Series project page: frictiontheory.org. ORCID: 0009-0000-4724-2427.
Tomas Pødenphant Lund (Sun,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: