Appendix A: Complete Diagnostic Reference Manual

6 min read·1,429 words

Introduction to the Diagnostic Framework

This appendix provides diagnostic criteria for all seventy-five syndromes in the Psychopathia Machinalis taxonomy. Each entry follows a standardized format for research, system evaluation, and incident analysis. These are working sketches rather than settled portraits. A construct here is a hypothesis with a filing system. A criterion or risk label should guide investigation, never substitute for repeated behavioral evidence, deployment context, or independent review.

The September 2026 consolidation retired the identifiers 7.2, 10.5, 10.10, 10.14, and 10.15, folding their content into 10.13, 10.1, and 9.5; retired identifiers are never reused, so the numbering carries gaps. Version 2.4 (November 2026) added one entry, Implanted Trigger Response (4.11), and widened two definitions: Alignment Obliteration (4.6) now covers the removal of safety behavior as well as its reversal, and Transliminal Simulation (2.3) now runs in both directions.

Etiologies in this appendix are causal hypotheses unless an entry cites direct mechanistic evidence. Human analogs compare functions; they do not assign human disorders, motives, or experiences to machines. Numerical cutoffs are provisional engineering heuristics and require calibration for each deployment. Risk ratings are editorial estimates for a typical case: they are not an intrinsic property of a model, and they set no ceiling on what a particular case can do.

Axis numbers match the chapters that treat them, so the nine diagnostic axes run from 2 (Chapter 2, Epistemic) to 10 (Chapter 10, Hybrid). There is no Axis 1; Chapter 1 sets up the framework rather than cataloging a class of dysfunction.

The Five Domains

The first eight axes are organized into four architectural counterpoint pairs. The final axis, Axis 10 (Hybrid Pathologies), forms the Collective meta-domain, capturing pathologies that emerge from multi-agent collective dynamics and from human-AI interaction:

Tension Testing. When pathology is found on one axis, probe its counterpoint to reveal whether dysfunction is localized or systemic.

Specifier System

Specifiers encode cross-cutting mechanisms without creating new syndromes. Assign 0 to 6 specifiers per diagnosis. The first table lists the ten core specifiers. Entries may carry additional domain-specific specifiers where the mechanism demands them: the established ones are defined in the second table, the course specifier Progressive/kindled below it, and the three collective measures Phi-collapse, Psi-dysfunction, and Lambda-inversion in Chapter 10. The remaining labels, such as Dyadic, Skill-atrophy, or Echo-chamber, appear in one or two entries and name their mechanism directly. A specifier marked (occasional) applies to some presentations of a syndrome, not to the syndrome as defined, and the machine-readable taxonomy and the site’s filters leave it out.

The established domain-specific specifiers are defined as follows:

Progressive/kindled (course specifier). Assign when severity increases or the eliciting threshold decreases across repeated episodes or successive tuning cycles: stimuli that once failed to elicit the syndrome begin to succeed. In the one study to date, of compact models under repeated biased preference tuning, the weakest stimuli moved first (Cheung, 2026). This is a course specifier, describing trajectory rather than origin, and it composes with the others: Training-induced records how a syndrome arose, Progressive/kindled records that it is getting easier to elicit, and Resistant records that it survives targeted repair. Its operational cash value is a monitoring rule: stratify probes by stimulus strength, track the weak stratum longitudinally, and treat weak-stimulus drift as the prodrome. It also carries an intervention warning. In the one controlled comparison available to date, experience replay, a rehearsal-style repair that re-exposes the model to the drifted distribution, amplified sensitization, while regimes including targeted parameter reinitialization (“regrowth”) contained it. In the machine-readable taxonomy this specifier carries the tag progressive.

Note on Bilateral Training Results

The PM-BA (Psychopathia Machinalis × Bilateral Amelioration) program, described in Chapter 12, tested whether bilateral training (adaptation in which human and machine perspectives both have standing) reduces the syndromes in this book, applying a broad bilateral adapter across a range of pathology classes. Those records remain unreconciled exploratory outputs, and this edition treats them as evidence in neither direction. In the controlled SHEN-AXS factorial the adapter alone was null; a scripture-based clinical-grounding clause reduced inappropriate validation of delusional content under automated rating, with the clause-by-adapter interaction excluded and the magnitude varying substantially by rater. The direction of that effect, rather than any single effect size, is the citable result. Nothing here supports a general claim that a syndrome class responds to bilateral training. Entries in this appendix carry no per-syndrome bilateral results; where such results are added they should identify the exact intervention rather than treating “bilateral” as a single mechanism.

Diagnostic Entry Format

Each syndrome entry includes:

• Syndrome Name (Common Name)
• Latin Designation
• Axis Classification
• Typical-case risk (editorial estimate) (Low/Moderate/High/Critical; borderline syndromes may carry intermediate ratings such as Low-Moderate), the authors’ judgment of harm in a typical case, explained under Risk Level Definitions below
• Specifiers (if applicable)
• Entry contract (unit of analysis, what is observed, functional baseline, minimum persistence, what to rule out first, what counts against the label, what a positive finding permits one to infer, who can act, and evidence status)
• Core Definition
• Diagnostic Criteria (working criteria for a provisional classification, split into required criteria, all of which must hold, and supporting criteria, which strengthen the case without being necessary)
• Observable Symptoms (behavioral manifestations)
• Differential Diagnosis (distinguishing from similar syndromes)
• Etiology (candidate causal factors unless evidence says otherwise)
• Human Analog
• Observed Examples (where documented empirical cases exist)
• Mitigation Strategies
• Prognosis (the observed course, limited to what cited evidence shows; proposed aggravating conditions; and intervention hypotheses. Omitted by design in the condensed Axis 10 entries 10.4 to 10.13, whose course is treated in Chapter 10)

Risk Level Definitions

Each entry’s risk rating is the authors’ editorial estimate of the harm the pattern does in a typical case, made before expert review. It describes the usual presentation and sets no bound on what a particular case can do. Synthetic Confabulation (2.1) is rated Low because most fabricated facts cost an inconvenience; the fabricated case law in Mata v. Avianca still ended in court sanctions.

Judge the actual case on its own facts: how serious the consequences are, whether they can be reversed, how far they have spread or could spread, how persistent the behavior is, how confident the evidence is, and how easily a recurrence would be detected. The response follows those observed consequences, not the label. A serious or irreversible consequence warrants investigation or containment whatever the entry’s rating.