Section 47 of 440

PART 2B: THE MEASURED CAPABILITY STANDARD

Capability Standard

Status and scope

This is Titan’s conformant operational implementation of HCE constitutional semantics v1.0, using editorial release v1.0.2. It applies to seven families and sixteen separately assessed modules. It adds assessment history, maintenance semantics, declaration locking, conditional rating reliability and cross-module safety review without changing the frozen twenty-four-domain Core, historical compatibility contract or Bridge rules.

Titan records a bounded capability claim: demonstrated performance by a declared actor or human-in-system configuration on a declared task, under recorded conditions, constraints and supports, with stated evidence and currentness. It does not score identity, worth, diagnosis, general intelligence or unlimited potential.

HCE normative mapping

The objects below are mandatory for interpreting an operational result. The full machine-readable Capability Genome is maintained in the Titan registry; this book presents the reader-facing contract.

Table 10 — Capability Standard
HCE object	Titan implementation	Governing rule
System of interest	Individual, team, assisted person or human-technology configuration.	Name the configuration that actually produced the result; do not attribute a larger-system outcome to a smaller actor without separate evidence.
Capability claim	Module, task, outcome, conditions, constraints, supports and currentness.	The claim cannot exceed the declared boundary.
Formation mechanism	Practice, calibration, substitution, recovery, environmental design, augmentation or collective coordination.	A mechanism tag describes the route; it does not prove effectiveness.
Assessment stages	Direct performance, delayed retention, predeclared transfer, burden and safety.	Each stage requires its own result; one stage cannot stand in for another.
Maturity metadata	Decoded, functional, transferred, incorporated, automatic, non-interfering, persistent and adopted where relevant.	Maturity is evidenced metadata, not L6 and not an overall score.
Evidence lifecycle	Current, Maintained, Review due, Historical or Criterion not met.	Currentness is separate from L1-L5; later success never erases prior history.
Capability Genome	Versioned registry record for identity, claim boundary, formation, evidence, assessment, ecology, economics, maintenance and governance.	Schema conformance checks structure; it does not prove the underlying capability claim.

Every claim-critical record stores the HCE constitution version, Genome-schema version, semantic-validator version, module version, protocol version and criterion version. If any of those meanings change, the prior record remains readable and a migration or new assessment context is created. Silent semantic change is prohibited.

The Constitution, Maturity Vector and Capability Viability dimensions are non-compensatory. A safety, consent or legality failure cannot be averaged away by good performance. Missing transfer cannot be hidden by a strong direct score. High burden cannot disappear inside a global index. Titan therefore keeps the relevant dimensions visible instead of collapsing them into one number.

Research-layer boundary. Compendium v1.1 reconciles research identities and sources but promotes no record. Research Expansion v1.0 is planned and not executed. MLCA v0.2 is Candidate-only. None changes the operational contracts below.

Frozen contracts

Table 11 — Capability Standard
Contract	Revision 7L treatment
Frozen Core domains	Exactly 24
Public Core display	Non-scoring profile
Historical Revision 7K composite	Preserved only in imported history
New global or layer calculation	Suspended; no output
Historical weighting	Archived for compatibility; not emitted
Automatic bottleneck selection	Prohibited
Automatic Program recommendation	Prohibited
Operational Capability results	Separate task-specific records; never averaged

Measurement contract

Record not assessed separately from level 1.
Use the module-specific anchor, not a general impression.
Level 3 requires a recorded direct result.
Level 4 requires delayed retention and a predeclared transfer result.
Level 5 requires robust performance across defined constraints plus explanation, teaching or independent-assessment evidence as the module requires.
Record burden, support, adverse events and data coverage beside performance.
Record lifecycle state separately from L1-L5; maintenance is a verified relationship between comparable assessments, never a sixth level.
When a subjective human rating is claim-critical, lock the rubric and agreement rule before observation and retain the independent pre-adjudication ratings.

Generic level gates

Table 12 — Capability Standard
Level	Minimum interpretation
L1	Baseline: performance is inconsistent or requires substantial support.
L2	Supported: the trained task is reliable only under narrow or supported conditions.
L3	Functional: the defined direct-performance criterion is met consistently.
L4	Transfer: the gain is retained and appears on a predeclared novel task or context.
L5	Robust: performance holds across defined constraints and the method can be explained or taught safely.

Module-specific anchors control where stricter. Not assessed is blank, not L1 and not zero. L3 requires a recorded direct result. L4 requires delayed retention and a predeclared novel transfer result. L5 requires robustness across declared constraints and the module’s teaching, explanation or independent-verification gate.

Verification standard

Evidence mode records how the results supporting a level were verified. It is neither a truth score nor an evidence-quality grade. If required components use different modes, record the least independently verifiable mode that applies to every component. Support used during performance is a separate field.

Table 13 — Capability Standard
Mode	Meaning
Self-report	The person reports an experience, judgement or outcome without a contemporaneous result log, artifact, instrument record, witness or independent assessment.
Self-recorded	The person records task conditions and results or retains an artifact, but no instrument, witness or independent assessor verifies the claimed result.
Instrumented	A declared device, sensor or software instrument produces the relevant measurement. This does not by itself prove identity, calibration, correct setup or independent oversight.
Witnessed	An identified observer attests that they directly observed the relevant performance and conditions. This does not by itself establish assessor competence or independence.
Independently assessed	A qualified or authorised third party, or an externally governed assessment process, judges the performance against disclosed criteria independently of the person being assessed.

Level rules

Table 14 — Capability Standard
Level	Verification requirement
3	Requires a direct result and at least Self-recorded evidence.
4	Requires direct, delayed-retention and predeclared-transfer results and at least Self-recorded evidence; a module may require more.
5	Defaults to Independently assessed. Instrumented or Witnessed is allowed only through explicit module metadata; Self-report and Self-recorded never qualify.
1-2	Self-report may be used only where the module anchor permits awareness or supported practice.

Historical records with no mode remain visible as legacy records with verification mode not recorded. Titan never infers a mode from level, prose, support, timestamps or attached filenames.

Maintenance and currentness

L1–L5 remains the performance scale. Lifecycle is displayed beside the task-specific level and never creates L6 or a cross-domain output.

A level says what criterion was demonstrated. Lifecycle says whether the evidence is current and how the latest comparable event relates to prior evidence.

Table 15 — Capability Standard
Lifecycle display	Meaning
Current	The latest criterion-met assessment remains inside its declared review window.
Maintained	A linked, in-window retest still meets the exact same target level, criterion version and protocol version under the declared comparable conditions.
Review due	The review window has ended or is about to end; a new result is required for a current claim.
Historical	The result is preserved but no longer presented as current evidence.
Criterion not met	The latest comparable assessment did not meet the declared criterion; earlier achievements remain provenance, not current status.

Required history rule

Every assessment is an immutable historical event. Corrections or revalidation create a linked successor; they do not overwrite the prior result.
Maintained may be derived only when an in-window retest links to a valid prior assessment, uses the exact same target level, criterion version and protocol version, preserves declared material conditions, support and evidence mode, and still meets the criterion. A changed criterion or protocol starts a new assessment context and is not maintenance.
Each result records assessedAt and reviewDueAt or validUntil. Overdue evidence becomes historical or review due; absence of a retest is never labelled maintained.
Recovery describes a later criterion-met assessment after a criterion-not-met event. It does not erase the interruption or imply medical recovery.
A changed criterion or protocol creates a new assessment context and cannot be labelled maintenance under the exact-comparison rule.

Measurement error, criterion attainment, and what can be claimed

Every criterion in this book asks whether current performance meets a predeclared threshold. A longitudinal claim asks a different question: whether the difference from an earlier score is sufficiently large to be distinguished from measurement error. Titan records criterion attainment and detected change separately; neither is inferred automatically from the other.

That distinction is established in outcome-measurement science. Test–retest reliability describes consistency when the measured construct is expected to be stable. The standard error of measurement estimates score-level error, and an individual-level smallest or minimal detectable change (SDC/MDC) expresses a threshold above which a difference may be distinguished from measurement error at a declared confidence under a particular instrument, population and protocol. A difference below that declared threshold is recorded as “change not detected at the stated confidence under this instrument and protocol.” It must not be labelled improvement or decline from that pair of observations. This does not establish that no underlying biological, cognitive, behavioural or functional change occurred.

Minimal important change (MIC) addresses a different question: whether a difference is important to the person or another declared stakeholder. SDC/MDC and MIC are distinct but must be interpreted together. Detectability does not establish importance, and importance does not establish detectability. Where MIC is undefined, interpretation is explicitly indeterminate; where SDC/MDC exceeds MIC, the instrument may be too noisy to distinguish a change as small as the amount considered important.

Measurement properties are not portable labels. Reliability, error and responsiveness may differ by instrument version, score units, population, administration mode, rater, retest interval, permitted support and conditions. A published SDC/MDC may be applied only after an explicit applicability decision. Repetition can improve an estimate, but it does not validate an unknown measure, remove systematic bias or convert an unvalidated task into a valid one.

For Titan’s single-person use, the defensible outputs are bounded and separate: current criterion attainment; detected change at a declared confidence; change not detected; indeterminate change; and importance to the person or declared stakeholder. A result may meet an absolute criterion without proving improvement from baseline, and a meaningful lived change may remain undetectable with the selected instrument.

Table 16 — Capability Standard
Requirement	Rule
Measurement identity	State the task or instrument ID and version, score units, actor or population, administration conditions, permitted supports, rater or mode, and applicable language or accessibility route.
Reliability and error model	State test–retest reliability or record it as unknown. Where an error threshold is used, state the SEM model or formula, retest interval, and whether relevant systematic and random error have been addressed.
Detectability threshold	State the individual-level SDC/MDC, confidence level, source, and an explicit applicability decision for this population, protocol, score and use. A group-level threshold must not be relabelled as an individual threshold.
Change rule	A difference below the declared individual-level SDC/MDC is recorded as change not detected at the stated confidence under this instrument and protocol. It is not labelled improvement or decline from that pair, and it is not proof that no underlying change occurred.
Importance	Record MIC separately where a defensible value or person-declared threshold exists. Interpret MIC with SDC/MDC: detectability does not establish importance; importance does not establish detectability.
Criterion attainment	Absolute attainment may be recorded when the current result meets a predeclared criterion under the declared conditions and supports. It is a current-performance claim, not automatically a longitudinal improvement claim.
Promotion rule	Every promotion contract states whether it requires criterion attainment, detected longitudinal change, or both. A small difference from baseline does not invalidate current attainment unless the declared promotion rule requires demonstrated change.
Unknown-properties default	Where reliability, measurement error or applicability is unknown, one score pair cannot establish improvement or decline. Repeated observations may support a bounded trend estimate, but do not validate the measure.
State and practice effects	Record material practice, familiarity, fatigue, symptoms, assistance, motivation, device or model version, and context changes that could alter the score or its interpretation.
Uncertainty and non-assessment	Report uncertainty where available. Preserve invalid, inaccessible, contraindicated, interrupted, declined and not-required outcomes as reason-coded non-assessments rather than failures.

This layer is aligned with COSMIN’s separation of validity, reliability, measurement error and responsiveness rather than treating “validated” as one status. Titan does not claim COSMIN conformance and does not transfer measurement properties automatically from one instrument, population or protocol to another. Where a Titan task has a governed counterpart in the measurement literature, the applicable instrument-specific evidence takes precedence over a generic Titan rule.

EVD-MEASUREMENT-ERROR-COSMIN — Mokkink LB, Elsman EBM, Terwee CB. COSMIN Manual for Systematic Reviews of Patient-Reported Outcome Measures. Version 2.0. COSMIN; August 2024. Measurement error is interpreted by comparing SDC or limits of agreement with MIC; where MIC is not defined, the result is indeterminate.

EVD-MEASUREMENT-ERROR-RISK-OF-BIAS — Mokkink LB, Boers M, van der Vleuten CPM, Bouter LM, Alonso J, Patrick DL, de Vet HCW, Terwee CB. COSMIN Risk of Bias tool to assess the quality of studies on reliability or measurement error of outcome measurement instruments: a Delphi study. BMC Med Res Methodol. 2020;20:293. DOI: 10.1186/s12874-020-01179-5.

Capability decay and refresher scheduling

Maintenance classes set a review window. They do not say how fast a capability is expected to fall, and a fixed ninety-day review applied to everything will be too late for some capabilities and pointless overhead for others. Revision 7E adds a declared decay expectation beside the maintenance class.

Recording rule. When a module or record is declared, state one decay band and the evidence it rests on. A decay band is an expectation used for scheduling, never a measurement and never a score. If the retest contradicts the band, the band is wrong and is revised; the retest stands.

Table 17 — Capability Standard
Decay band	Scheduling consequence
D1 — slow	Well-consolidated procedural or embodied skill with continuing incidental use. Refresher at the maintenance-class interval; no interim probe needed.
D2 — moderate	Knowledge or rule-based skill used intermittently. Insert one low-cost retrieval probe at roughly half the maintenance interval.
D3 — fast	Complex multi-step procedures, emergency responses, and any capability practised only under instruction. Probe early and often; assume material loss well inside the maintenance window.
D4 — device- or support-dependent	Capability expressed through a device, decoder, prompt or assistant. Schedule a support-off or unaided retest as well as the ordinary refresher; decay of the person and drift of the support are different failures and must be measured separately.
D0 — undeclared	Default where no evidence supports a band. Treated as D3 for scheduling until declared. Undeclared is not the same as slow.

Non-use is the scheduling variable, not elapsed time. A capability exercised daily inside ordinary work does not need the same refresher as the identical capability left untouched for a quarter. Where the Engine can observe use, the interval runs from last use; where it cannot, it runs from last verified retest and the difference is recorded.

Boundary. A decay band never raises or lowers a capability level, never substitutes for a retention test, and never licenses a claim that a capability is retained because the band says it should be. It schedules the check. Only the check settles the claim.

Evidence. A meta-analysis of 1,344 effect sizes from 457 reports found that performance decrements grew by approximately 0.08 standard-deviation units per month for accuracy-based measures and 0.06 for speed-based and mixed measures. Without use, roughly half of the gains from initial acquisition were lost after about 6.5 months for accuracy, 13 months for speed and 11 months for mixed measures. Task type, task complexity, intermittent performance opportunities and task instructions were identified as possible moderators. See EVD-SKILL-DECAY-2025.

Band anchoring. Where no capability-specific evidence exists, use those pooled estimates as the default: accuracy-dominant capabilities are treated as D2 or faster, speed-dominant capabilities as D1 or D2. These are population averages across heterogeneous tasks, not predictions about any individual or any particular skill, and they are a starting point for scheduling only. The retest, not the estimate, settles whether a capability is retained.

Candidate-architecture boundary. Spaced-repetition systems can estimate item-level difficulty, stability and retrievability from review histories, but the useful data volume and model validity depend on the implementation, version, item type and use case. There is no universal scientific threshold of one thousand reviews at which an individual model becomes valid or superior. Current FSRS implementations do not impose that as a general optimisation minimum; earlier software versions used different gates. Titan therefore makes no fixed-observation claim and no promise that personal fitting will outperform a simpler schedule.

Candidate architecture from spaced repetition. Difficulty, stability and retrievability provide a useful vocabulary for item-level declarative recall. They are not yet a validated universal model of procedural, perceptual, physical, relational, organisational or human–technology capability decay. D0–D4 therefore remain Titan’s declared scheduling fallback. A DSR-inspired capability model may be promoted only after preregistered comparison with simpler baselines across capability classes, measuring calibration, prediction error, retention decisions, maintenance burden, accessibility, failure costs and support drift.

Implementation provenance — The official FSRS tutorial states that Anki 24.06 and later has no universal minimum review count for parameter optimisation; Anki 24.04 used 400 reviews and older versions used 1,000. This is software-version provenance, not evidence that DSR generalises to Titan capability classes. Source: open-spaced-repetition/fsrs4anki, tutorial.md, accessed 26 July 2026.

The airline industry has gone further. Evidence-based training programmes select and prioritise recurrent training topics from operational data specifically to minimise competence decay over time, rather than repeating a fixed manoeuvre set. Titan’s decay bands are a smaller, self-directed version of an approach that safety-critical industries already run at scale. This is stated here because a reader is better served knowing the idea is established and tested than believing it is new.

The honest summary is that the decay bands here are a simplified, self-directed version of a modelled discipline that safety-critical professions have practised since the 1980s. That is not a reason to remove them — a reader has no access to a unit training schedule — but it is a reason to state that better evidence exists, and that anyone wanting precision should look to that literature rather than to this book.

NATO human-factors work has since developed the approach in two directions: statistical models that set refresher intervals at unit level, and analytics that build a personalised retention profile so that refresher training reaches an individual just in time. That second direction is substantially what the Titan Engine proposes to do. The same literature identifies dozens of factors influencing skill fade, spanning person-related, task-related and contextual categories, and makes a point this framework currently lacks: within a single composite task, different components decay at different rates, and a slowly-decaying component can carry recall of a faster-decaying one.

Defence went first, and by four decades. A skill-decay prediction model called the User Decision Aid was developed for United States military trainers in 1984 to answer three questions directly: which tasks are most likely to be lost to decay, what proportion of operators will still perform a task correctly after a given number of months without practice, and when and how often sustainment training should be run. It was applied across trades including vehicle mechanics, radio operators, quartermasters, combat engineers, field medics and air-defence crews. Subsequent validation found it generally over-predicted decay — which is itself the kind of finding this framework says should be preserved.

Prior art, and it is substantial. Decay-informed refresher scheduling is not new. Civil aviation has operated recency rules for decades: a pilot must have completed a minimum number of takeoffs and landings within the preceding ninety days to carry passengers, with separate recency requirements for night operations. The regulatory structure is the same one used here — the certificate does not expire, but currency lapses without use, and lapsed currency suspends the privilege rather than erasing the qualification. That is exactly Titan’s rule that maintenance is a lifecycle state and never a sixth level.

Declaration locking

Bind criteria and tests before observation without claiming that a local timestamp proves external pre-registration.

Exact protected-field set

The following fields, in this registry-defined set, are the complete declaration payload covered by TITAN-CANONICAL-JSON-1. The stored digest is deliberately excluded from the hashed object.

id
moduleId
revision
targetLevel
task
conditions
baselinePlan
directCriterion
retentionPlan
transferPlan
evidencePlan
burdenPlan
stopRule
protocolVersion
criterionVersion
reviewDueDays
historicalAfterDays
createdAt
lockedAt
supersedesDeclarationId
canonicalization
digestAlgorithm

Lock, digest and linkage rules

Table 18 — Capability Standard
Control	Rule
Lock	A locked declaration is append-only. A change creates a new declaration with supersedesDeclarationId and keeps the prior declaration and digest visible.
Digest	TITAN-CANONICAL-JSON-1: the requiredFields above, including canonicalization=RFC8785 and digestAlgorithm=SHA-256, are serialized as RFC 8785 canonical I-JSON and hashed as UTF-8 bytes; the digest is encoded as exactly 64 lowercase hexadecimal characters and is not itself part of the hashed object.
Result linkage	Every assessment result created under a declaration references its declarationId and digest. An unlinked or post-hoc result is labelled undeclared and cannot satisfy a predeclaration gate.
Time boundary	lockedAt is a device-reported timestamp unless accompanied by a trusted signed or externally anchored receipt. A local timestamp plus a recomputable local hash does not prove intention before observation.
Signature boundary	A signed declaration receipt can prove key-and-bytes integrity and, with an approved time authority, receipt time. It does not prove identity, correct observation, truth or capability.

A locked local declaration is useful discipline. Its digest detects a mismatch between the stored protected fields and stored digest at verification time; a person who edits the fields and recomputes the digest can defeat that local check. It becomes stronger evidence of intention-before-observation only when an approved receipt authority signs or externally anchors the declaration digest at lock time. That service remains disabled in the Revision 7N candidate.

Rubric-scoring reliability

Applies when. A subjective human-coded rating materially determines an L4 or L5 claim, or a module or exemplar explicitly requires independent rubric scoring.

Waiver. Not required where a fixed answer key, calibrated instrument or externally governed assessment determines the claimed result without subjective coding.

Freeze the rubric, examples, rater eligibility, agreement metric, threshold and disagreement response before ratings begin.
Have at least two eligible raters score independently without seeing the other rating.
Preserve every pre-adjudication score and report raw agreement plus the predeclared statistic.
Use weighted kappa for ordinal categories, an appropriate intraclass correlation for continuous scores, or Krippendorff’s alpha when multiple raters or missing ratings require it.
If the threshold is not met, retrain or clarify the rubric and collect a new independently rated set; never silently average disagreement away.

Boundary. Agreement does not establish rubric validity, truth or general transfer. It establishes only that the declared coding rule was applied consistently enough for the claimed use.

Cross-module safety review and professional handoff

Make repeated burden, adverse events, near misses and stop-rule actions visible across modules without diagnosing their cause.

Safety events are append-only. The initial event has previousSafetyEventId=null and cannot begin as Resolved. Monitoring, resolution or reopening is appended as a structured linked successor; the earlier status and event record are never edited or deleted.
The review surface may show recurrence, unresolved chains and burden by module or category. Only each valid chain head contributes to the open/monitoring count, so a resolved predecessor is not left counted open. The review must remain descriptive, avoid causal or diagnostic inference, and list Stopped session and Emergency response records individually even after a linked resolution rather than concealing them inside an average.
Store the least sensitive description needed for action. Professional handoffs are selective, consent-driven exports rather than automatic disclosure.
A handoff is selected by the person, limited to relevant modules and dates, and labelled educational and non-diagnostic. It is not clearance, certification or an instruction to a professional.
A coach export excludes health or adverse-event details by default; a clinician handoff includes only details the person deliberately selects.

Population-reference boundary

Population ranks and percentiles never determine a Titan level, badge, worth or global comparison. A validated instrument-specific reference may be stored as non-scoring safety, referral or criterion-selection metadata only for the population and purpose it supports; it never replaces predeclared direct, retention and transfer evidence.

Capability Acquisition Layer

A Capability Genome records not only what a person or human-system configuration can do, but how the capability is acquired. The acquisition mechanism does not replace the direct outcome, retention test, transfer test, burden record or safety gate. It prevents distinct learning processes from being collapsed into a vague claim of enhancement.

Registry rule. Store one primary acquisition mechanism and, only when necessary, one secondary mechanism. “Mixed” is not an acceptable substitute for identifying the dominant process. The field may be revised when evidence shows that the original mechanism was wrong.

Table 19 — Capability Standard
Controlled value	Operational meaning	Illustrative examples
latent_channel_recruitment	Brings an existing but weakly conscious signal into reliable use.	Polarization-pattern perception.
perceptual_calibration	Improves task-specific resolution within an existing sensory channel.	Absolute pitch; visual hyperacuity.
perceptual_remapping	Learns a changed relationship between a signal and the environment.	Altered auditory spatial cues.
cross_modal_substitution	Translates information usually carried by one sense into another.	Image-to-sound; distance-to-vibration.
sensory_augmentation	Adds a continuous environmental variable not normally available in that form.	Vibrotactile north; synthetic self-motion cue.
procedural_compilation	Turns an explicit algorithm or rule set into rapid, reliable performance.	Calendar calculation; mental arithmetic procedures.
physiological_self_regulation	Uses feedback to learn reproducible influence over a measured physiological variable.	Pupil or cardiovascular biofeedback.
body_schema_expansion	Incorporates an extra tool or effector into action planning.	Robotic thumb; extra virtual limb.
reciprocal_adaptation	The person and decoder, AI or device adapt to one another over time.	Adaptive BCI; shared-control system.
state_gated_cognition	Uses a naturally occurring state associated with different information processing.	N1 insight capture; verified lucid-dream interaction.
collective_emergence	Produces a capability at group level through preserved independence and coordination.	Calibrated aggregation; team orchestration.
error_management_acquisition	Trains a skill under deliberately errorful and varied conditions so that performance survives novel cases rather than only familiar ones.	Error-management training; productive failure; adaptive expertise.

Capability Maturity Vector

Meeting a task criterion does not necessarily mean that a signal, tool or procedure has become automatic, embodied or durably incorporated. Where relevant, Titan records a separate maturity vector beside the L1-L5 capability level. It never replaces the level or raises it by itself.

Recording rule. Each field is Not tested, Not met, Met or Uncertain, with date, task, evidence mode and version. Subjective familiarity may support an experience note but cannot establish incorporation, automaticity or persistence.

Table 20 — Capability Standard
Field	Question	Minimum evidence
decoded	Can the person consciously interpret the signal above chance?	Blinded or randomized signal-only task.
functional	Does it improve the declared task?	Direct criterion against baseline.
transferred	Does it work on a predeclared novel stimulus or context?	Module transfer test.
incorporated	Is it combined appropriately with existing information?	Combined-cue benefit, conflict test or reliability reweighting.
automatic	Does useful performance persist under a secondary task?	Predeclared dual-task comparison.
non_interfering	Is the native skill or original mapping preserved?	Original-function retest and adverse-event review.
persistent	Does the ability remain after feedback, equipment or intensive practice is removed?	Device-off or feedback-off delayed retest.
adopted	Does the user voluntarily use it when useful?	Real-use log with opportunities, use and non-use reasons.

Claim rule. The practical claim remains bounded by the weakest required evidence. “Incorporated,” “embodied,” “automatic” or “new sense” language requires the corresponding objective test; it is never inferred from novelty or subjective intuition.

Relationship to existing classification systems

Titan did not invent most of what it does, and saying so plainly is worth more than the alternative. Several of its central mechanisms have mature prior art, and the framework is more credible for citing it than for quietly re-deriving it.

Declared tasks. Stating the task before assessing it is standard practice elsewhere. The World Health Organization's International Classification of Functioning, Disability and Health classifies actions, tasks and areas of life; the Common European Framework of Reference describes language ability through can-do statements at six levels; and competency-based medical education assesses entrustable professional activities, which are observable task units. Titan follows that practice rather than establishing it.

Unaided versus supported. The distinction Titan draws between what a person can do alone and what they do with a device, another person or an AI system is the ICF's capacity and performance qualifier pair, in use internationally since 2001. The ICF also codes environmental factors as barriers or facilitators, which is what Titan records as supports and constraints. Titan adopts this distinction; it does not originate it.

Graded independence. Titan's L1-L5 scale, in which a level records what is demonstrated and under what support, is structurally close to the entrustment-supervision scales used in competency-based medical education, where a five-point scale records how much supervision a learner needs to perform an activity safely. That literature is two decades old and its findings apply here.

One of those findings applies with extra force. Entrustment in professional education is judged by a supervisor who observes the learner. Titan has no supervisor. A self-declared level rests on one observer who is also the person being assessed, and every reliability concern documented in the professional-education literature should be read as a lower bound on the problem here, not an upper one. That is why declaration locking, predeclared criteria, verification mode and the requirement to record support honestly are not bureaucratic decoration. They are the only substitutes available for a second observer.

Evidence grading. Rating the strength of evidence behind a claim is established methodology, most prominently through the GRADE approach used in clinical guideline development. Titan's evidence grades serve a narrower purpose and are less formally defined. They should be read as a working scale for a self-tracking framework, not as an equivalent to a clinical evidence appraisal.

Titan’s grades are not GRADE grades and should not be presented as equivalent. GRADE rates the certainty that an effect estimate is adequate to support a recommendation, and does so through a defined process of assessing risk of bias, inconsistency, indirectness, imprecision and publication bias. Titan’s grades rate how far a capability has been shown to be trainable, retained and transferable. Where a Titan record touches health, the GRADE-rated clinical literature governs and the Titan grade is subordinate to it. A Titan grade of Strong never licenses a health claim that the clinical evidence does not support.

A note on the word capability

In this book, a capability is a demonstrated, reproducible performance on a declared task under declared conditions.

That is not what the word means in the capability approach developed by Amartya Sen and Martha Nussbaum, which is influential in development economics, philosophy, disability studies and public policy. There, a capability is a person's real freedom or opportunity to achieve valued ways of being and doing; the achievements themselves are called functionings. By that vocabulary, what Titan calls a capability is closer to a functioning, and what that tradition calls a capability is a question about opportunity and circumstance that this book does not address and should not be read as addressing.

The distinction matters because the stakes differ. The capability approach asks what freedoms a person must have for a life to be dignified. Titan asks whether a particular task performance can be demonstrated, retained and transferred. Nothing in this book enumerates what a person needs in order to flourish, and nothing in it should be used to argue that someone lacks a capability in the philosophical sense because a task criterion was not met.

Registry namespace and identifier prefixes

Every Titan record carries a prefix that states what kind of claim it is. The prefix is part of the identifier, not decoration, and the distinctness gate is applied against it. Reading a record without reading its prefix is the commonest way to mistake a research watch item for an operational capability.

MOD- denotes an operational capability module with a canonical Genome record. CAND- denotes a Candidate Lab record under bounded test. OBS- denotes a Research Observatory watch item, which can never alter the Core profile or an operational level. EX- denotes a worked exemplar inside a module. EVD- denotes a verified evidence record. TFA- denotes a Capability Frontier Audit candidate that has not yet been adjudicated into any of the above. A separate governance namespace, HCE-TITAN-..., identifies the provenance record behind a module. A governance record identity is not a capability record and never counts as one; it exists so that a claim can be traced to the versioned source that governs it.

Deprecation notice. Seven Candidate Lab records carry legacy MOD- identifiers issued before this rule existed: Focus & Interruption Management, Adherence & Habit Formation, Stress Recovery & Performance Under Pressure, Information Verification, First-Aid & Emergency Readiness, Sleep Regularity & Recovery, and Functional Physical Capacity. These are candidates, not operational modules. The identifiers are retained because the evidence history is append-only and renaming would break existing linkage; each record is instead labelled at the point of use. They must not be counted among the sixteen operational modules, and a reader who counts MOD- strings in this book will arrive at twenty-three rather than sixteen for that reason alone.

Capability families

Table 21 — Capability Standard
Family	Purpose	Modules
Perception, Spatial Cognition & Sensorimotor Navigation	Extract useful information from the environment, construct spatial models and convert perception into safe action.	3
Learning, Memory, Creativity & Metacognition	Acquire, retain, retrieve, transfer, generate and accurately monitor knowledge and skill.	4
Quantitative, Decision & Causal Intelligence	Estimate uncertainty, choose actions, test causal explanations and learn from resolved outcomes.	2
Communication, Argument & Negotiation	Listen, explain, teach, test claims, negotiate and repair understanding without sacrificing truth or consent.	2
Collaboration & Collective Intelligence	Combine independent information, expertise, dissent, coordination and shared memory into better group performance.	1
Making, Tool Use & Operations	Diagnose, make, maintain and operate physical or procedural systems with verification, fallbacks and safe escalation.	2
Human–Technology Augmentation	Select and govern assistive or AI tools so the combined system improves outcomes without losing safety, ownership or skill.	2

Core-to-capability coverage

The Core and the Capability Standard are two lanes into the same architecture, and until this revision nobody had checked whether they cover the same ground. They do not.

The frozen Core has twenty-four domains across Hardware, Operating System and Applications. Revision 7N does not weight or combine those domains; Revision 7K weighting is retained only as historical compatibility data. The sixteen operational modules sit in seven families, and every one of those families is an Applications-layer family. Eighteen of the twenty-four Core domains therefore have no operational capability module — no declared task, no retention test, no transfer test, no L1–L5 route.

Table 22 — Capability Standard
Core layer	Domains	Operational modules
Hardware	Cardiovascular capacity; strength; metabolic health; mobility and structure; body composition; hormonal health; culinary and nutritional capability	None
Operating System	Sleep quality; autonomic balance; stress resilience; recovery capacity; breath control; dopamine and reward balance	None
Applications	Cognitive; social; physical skills; creative; epistemic discernment; information sovereignty	All sixteen
Applications, uncovered	Emotional; financial; archetypal presence; relationship depth and repair; legacy and mortality awareness	None

What this means in practice. A reader in the Core lane may record context across all twenty-four domains without receiving an overall score. A reader in the capability lane can currently build a declared, retained, transfer-tested claim only in the cognitive and social portion of the Applications layer. The lanes are complementary but measure different things and remain separate.

This is not a case for reopening the Core. The domains are right; what is missing is the measured route into two-thirds of them. Two families are therefore declared below as reserved. Declaring a gap and leaving it visibly empty is preferable to a coverage claim that the registry does not support.

Reserved capability families

Two families are declared and deliberately unpopulated. A reserved family has a name, a purpose, a family boundary and named module slots, but no operational modules. It exists so that the gap is governed rather than invisible, and so that the promotion route into it is stated in advance.

No promotion has occurred. Promotion remains a governance act requiring verified evidence, completed safety and accessibility review, validated data fields, adversarial metric tests, migration tests and a new registry release. Nothing below is scoreable, and a reserved slot must not be treated as a module in waiting whose outcome is already decided.

Family 8 (reserved) — Physical & Metabolic Capability

Purpose. Produce, sustain and recover declared physical work, and maintain the biological substrate on which every other family depends.

Family boundary. No aesthetic claim, no body-composition target, no athletic-potential claim, no diagnosis, and no exercise or rehabilitation prescription. A physical capability record would state a task a person can perform, not a body a person should have. This boundary is stricter than the others in the book because the surrounding culture is relentless here, and a measured physical claim is unusually easy to convert into a claim about someone's worth.

Table 23 — Capability Standard
Reserved slot	Core domains served	Nearest existing record and what blocks promotion
Functional physical capacity	Cardiovascular capacity; strength; mobility and structure	Candidate Lab record already exists. Blocked on: instrument pack per accepted test; proof that equipment, assistance and practice effects cannot inflate a result; witnessed evidence for higher levels; clinical and accessibility review covering falls, fractures, cardiac, respiratory and neurological risk; a transfer rule that measures a second function directly.
Metabolic and nutritional function	Metabolic health; body composition; culinary and nutritional capability	No record exists. Blocked on: finding any declarable task that is not a biomarker target. Note that culinary capability is a genuine skill with an observable output, and is the most promotable element here.
Endocrine and life-stage function	Hormonal health	No record exists, and may never be appropriate. Hormonal health is largely a clinical matter, not a trainable capability, and belongs with preventive care rather than in a capability registry.

Family 9 (reserved) — Regulation, Recovery & Attention

Purpose. Influence one's own physiological and attentional state reliably enough to change performance on a declared task, and recover afterwards.

Family boundary. No claim of emotional mastery, stress-proofing, autonomic control as a general trait, or resilience as a personal quality. No treatment for any mental or physical health condition. A regulation capability record would state that a person can reach a measurable state under declared conditions and use it for a declared purpose. It would not state that they are calm, disciplined or resilient.

Table 24 — Capability Standard
Reserved slot	Core domains served	Nearest existing record and what blocks promotion
Sleep regularity	Sleep quality; recovery capacity	Candidate Lab record exists. Blocked on: validated index implementation; proof that missing nights or reduced sleep opportunity cannot improve the score; an interpretable minimum-change policy; a no-wearable accessible route; independent sleep-clinical review of the insomnia, apnoea, shift-work and mania exclusions; monitoring for orthosomnia as an adverse outcome.
Stress recovery and performance under pressure	Stress resilience; autonomic balance	Candidate Lab record exists. Blocked on: task-specific delayed-retention evidence; an adverse-event framework for non-clinical use; qualified clinical approval of the boundary between bounded performance practice and treatment.
Attention and interruption management	Dopamine and reward balance	Candidate Lab record exists. Blocked on: a recent field intervention with objective real-work outcomes and delayed follow-up; feasibility of the candidate threshold without quality loss.
Emotional regulation	Emotional (Applications layer)	No record exists. This is the largest single hole in the registry: the book contains extensive emotional-regulation content but no measured capability route. Blocked on: finding a declared task that is not a self-report questionnaire, and a safety boundary that keeps the record clear of clinical territory.
Interoceptive discrimination	Autonomic balance; breath control	Observatory entry exists. Blocked on trainability evidence with retention. Breath control specifically has no route at all and is named once in this book.

Five Core domains remain with neither a module nor a reserved slot: financial, archetypal presence, relationship depth and repair, legacy and mortality awareness, and information sovereignty. Each has substantial content in Part 2 and none has a measured route. Whether they should have one is an open question rather than an oversight. Some human capacities matter enormously and measure badly, and a framework that admits this is more trustworthy than one that invents a criterion in order to fill a cell.

Module registry completeness

Sixteen operational modules are in force and all sixteen carry a full HCE Genome panel. Revision 7F recorded the Spatial Simulation panel as pending. That was wrong: the panel existed and only its caption was missing. The error came from a completeness check that counted captions rather than records, and it is recorded here rather than quietly corrected, because a registry that misreports its own completeness is a worse defect than an incomplete registry.

Table 25 — Capability Standard
Module	Canonical Genome record
MOD-SENSORY-DISCRIMINATION	Present, record 1.0.0
MOD-SPATIAL-SIMULATION	Present, record 1.0.0 (caption added in Revision 7H; panel unchanged since 7D)
MOD-SENSORIMOTOR-RESILIENCE	Present, record 1.0.0
MOD-LEARNING-TRANSFER	Present, record 1.0.0
MOD-METACOGNITIVE-CALIBRATION	Present, record 1.0.0
MOD-CREATIVE-COGNITION	Present, record 1.0.0
MOD-PROSPECTIVE-MEMORY	Present, record 1.0.0
MOD-FORECASTING	Present, record 1.0.0
MOD-CAUSAL-EXPERIMENTAL	Present, record 1.0.0
MOD-ARGUMENTATION	Present, record 1.0.0
MOD-TEACHING-NEGOTIATION	Present, record 1.0.0
MOD-TEAM-ORCHESTRATION	Present, record 1.0.0
MOD-TOOL-REPAIR	Present, record 1.0.0
MOD-OPERATIONAL-SYSTEMS	Present, record 1.0.0
MOD-HUMAN-AI-COMPLEMENTARITY	Present, record 1.0.0
MOD-ASSISTIVE-FIT	Present, record 1.0.0

A pending Genome record does not suspend the module, but it does mean the module's claim boundary, safety gates and maintenance class are governed by its summary record until the panel is issued. Declaring the gap is preferred to a silently incomplete registry.

Governance

Table 26 — Capability Standard
Control	Rule	Measure
Safety & Clearance	Risk screens, professional boundaries and stop conditions route every protocol before performance.	Clearance coverage, stop-rule compliance, adverse events and near misses.
Readiness & Dose Routing	Readiness is a transparent self-report pacing heuristic, not a validated diagnostic or capability score.	Dimension coverage, selected pace, planned versus completed dose and symptom response.
Accessibility & Adaptation	Preserve the intended function through an equivalent lower-risk pathway; never penalise an assistive route.	Constraint coverage, substitution success, burden and sustainable adherence.
Evidence Maturity & Claim Scope	The practical claim cannot exceed the population, task, outcome, transfer and safety supported by its evidence.	Resolved evidence links, claim-limit coverage, review date and downgrade actions.
Data Quality, Privacy & Retirement	Store only useful data, label missing coverage, protect exports and retire measures or protocols that do not earn their burden.	Data coverage, verification, privacy exposure, burden and keep/adapt/pause/retire decision.

Accessibility and adaptive routes

Table 27 — Capability Standard
Constraint	Equivalent route
Limited mobility	Use seated, supported or clinician-approved variants and score the intended function.
Sensory loss	Use accessible cues, assistive technology and qualified rehabilitation support.
Neurodivergent needs	Reduce sensory load, externalise sequencing and use predictable, interest-aligned blocks.
No specialist equipment	Use low-cost, software-free or household equivalents before buying tools.
Under 20 minutes a day	Use one active module, one direct measure and one weekly review.
Limited budget	Prioritise free fundamentals, public resources, repair, borrowing and low-burden trials.
Caregiving or unpredictable schedule	Use interruptible sessions, rolling weekly targets and recovery-aware rescheduling.
Pain or fluctuating fatigue	Use pacing and stop rules; replace intensity with tolerable consistency and qualified guidance.

An accommodation is recorded, not penalised. Preserve the intended function, limit the claim to the route actually tested, and apply the same stop rules and professional gates.

Operational modules

How to read the HCE Genome panels. Each panel is a compact publication view of the module’s canonical machine-readable record. Identity and status say what is being governed; purpose and outcome define the intended function; the claim boundary prevents the result from expanding beyond the tested actor, task, conditions, support and time. Formation design records how capability is expected to develop. Evidence and verification separate direct performance from delayed retention and predeclared transfer. Safety, consent and agency remain hard limits. Dependencies and ecology expose prerequisites, tools and typed relationships with other modules. Economics records burden and opportunity cost. Maintenance and retirement state how a claim stays current, lapses or is withdrawn. Governance and provenance identify the versioned source. These panels describe the module contract; they do not report that any individual has achieved it.

Revision 3 editorial rule. The module-native governing claim, claim boundary and evidence contract now take precedence over any broader legacy shorthand elsewhere in the book. This precedence narrows interpretation; it does not silently raise a score, relax a gate or create a new capability claim.

Perception, Spatial Cognition & Sensorimotor Navigation

Purpose. Extract useful information from the environment, construct spatial models and convert perception into safe action.

Family boundary. No claim of a new biological sense, unsupervised hazard navigation or universal sensory enhancement.

Sensory Discrimination & Substitution

Versioned exception — human click-based echolocation. Human click-based echolocation is not an authorised operational worked exemplar under this module. The module's measures, protocol, level anchors, retention tests and transfer tests do not apply to it. PWR-078 remains a frozen research candidate on publication hold; the module retains conceptual/read-only non-echolocation scope only.

HCE Genome panel · MOD-SENSORY-DISCRIMINATION · canonical registry record 1.0.0

HCE Genome panel · MOD-SENSORY-DISCRIMINATION · canonical registry record 1.0.0
HCE GENOME FIELD	CANONICAL MODULE CONFIGURATION
Identity & status	Sensory Discrimination & Substitution · family Perception Spatial Cognition Sensorimotor Navigation · actor scope Individual · status Operational.
Purpose	Extract task-relevant information from a defined sensory or substitution channel and use it accurately under controlled, safe conditions.
Outcome & claim mode	Accuracy, discrimination threshold or localisation error on a predeclared stimulus set, with chance level and scoring method disclosed. Claim mode: Confirmatory; formation route: Acquisition.
Claim boundary	Population: Adults or configured teams for whom the declared task and route are appropriate. Conditions: predeclared task; versioned protocol and criterion; declared support and constraints. Excludes: a new biological sense; safe independent navigation; general sensory superiority; diagnosis or treatment of sensory loss; plus 1 more.
Formation design	Primary mechanism: Perceptual Calibration. Secondary: Cross Modal Substitution. Patterns: calibration; substitution; scaffolding and fading; cue reliability testing.
Evidence & verification	Evidence grade: Moderate; lifecycle: Not Assessed; mode: Instrumented. Baseline: Record performance before the formation protocol using the same or a matched scoring rule. Direct: Accuracy, discrimination threshold or localisation error on a predeclared stimulus set, with chance level and scoring method disclosed. Retention: Repeat the same criterion after at least seven days without feedback-rich practice. Transfer: Meet a predeclared criterion on novel, equally safe stimuli, positions, frequencies or channel mappings.
Failure & interference	Failure modes: stimulus memorisation; criterion gaming; feedback dependence; cue over-weighting; native-sense interference; unsafe generalisation; sensory overload. Interference plan: Measure and review: stimulus memorisation, criterion gaming, feedback dependence, cue over-weighting, native-sense interference
Safety, consent & agency	Risk: Variable. Voluntary participation with accessible withdrawal and no penalty Professional boundary: Never train blind or transformed-sound navigation near traffic, stairs, water, heights, moving machinery or other hazards. Sensory-loss rehabilitation belongs with qualified support. Stop for persistent visual or hearing change, severe headache, nausea, disorientation or distress. Agency boundary: Capability support must not erase consent, accountable ownership or an appropriate fallback route.
Dependencies & ecology	Prerequisites: attention to task; basic instruction comprehension. Tools: None declared. Relationships: enables → MOD-SPATIAL-SIMULATION; synergistic → MOD-ASSISTIVE-FIT.
Economics & burden	Equipment: Variable: many calibration tasks need only software or household materials; substitution and augmentation may require specialist devices. Opportunity cost: Training and testing time; sensory fatigue; equipment and calibration burden. Maintenance cost: M2 — review after 30–90 days where continued use matters; device-mediated channels also require calibration and device-off checks.
Maintenance & retirement	Class M2; review due 90 days. Criterion: Repeat the declared direct criterion and review transfer, burden, safety and context change. Retirement triggers: criterion no longer met; unacceptable burden or safety event; invalid measure; loss of practical utility; plus 1 more.
Governance & provenance	Record HCE-TITAN-MOD-SENSORY-001 · schema 0.2.0 · record version 1.0.0 · authority Titan · review 2026-10-24 · history append only with tombstones.

Table 29 — Capability Standard
Field	Value
ID	`MOD-SENSORY-DISCRIMINATION`
Status	continued
Evidence	Moderate
Risk	Variable

Governing capability claim. Extract task-relevant information from a defined sensory or substitution channel and use it accurately under controlled, safe conditions.

Claim boundary. This record does not establish a new biological sense; safe independent navigation; general sensory superiority; diagnosis or treatment of sensory loss; performance outside the declared stimulus and safety envelope.

Evidence contract. Direct: Accuracy, discrimination threshold or localisation error on a predeclared stimulus set, with chance level and scoring method disclosed. Retention: Repeat the same criterion after at least seven days without feedback-rich practice. Transfer: Meet a predeclared criterion on novel, equally safe stimuli, positions, frequencies or channel mappings. The claim remains task-, route-, support- and version-specific; structural conformance does not prove empirical validity.

Editorial reconciliation. The canonical Genome record governs interpretation where older shorthand could be read more broadly. Existing protocols, measures, level anchors, safety gates and worked examples remain in force unless a later versioned change explicitly replaces them.

Measures

Table 30 — Capability Standard
Measure	Operational definition
Direct	Accuracy or threshold on a defined stimulus set.
Retention	Repeat the same test after at least seven days.
Transfer	Accuracy on a predeclared novel but equally safe stimulus set.
Burden	Time, irritation, fatigue, disorientation or distress.

Protocol

Choose one channel and one controlled task.
Establish a baseline with objective feedback.
Practise briefly while rested.
Repeat the test after a delay.
Test a novel safe stimulus.
Stop for disorientation, irritation, hearing change or distress.

Level anchors

Table 31 — Capability Standard
Level	Module-specific anchor
L1	Baseline accuracy is unstable or near chance on the defined task.
L2	Accuracy improves on the practised set with immediate feedback or support.
L3	Meets the predeclared accuracy threshold on the practised set in two sessions.
L4	Retains the threshold and meets the novel-stimulus transfer criterion.
L5	Performance remains reliable across defined safe variations and the method can be explained without overstating its scope.

Guardrail. Never train blind navigation near traffic, stairs, water, heights or moving machinery. Sensory-loss training belongs with qualified rehabilitation support.

Worked exemplars

Legacy record: Human click-based echolocation — publication hold

ID. EX-ECHOLOCATION — RETIRED_FROM_OPERATIONAL_USE__LINEAGE_ONLY

Status and safety disposition. This historical worked exemplar is retained as lineage only. It is not a protocol, training method, assessment path, level route or navigation recommendation. Echolocation may complement established mobility methods; this entry does not establish safe independent navigation and does not advise replacing a cane, guide dog, live assistance, or qualified orientation-and-mobility support.

Table 32 — Capability Standard
Test	Definition
Direct	Size or orientation discrimination accuracy, or virtual-maze completion time, completion rate and collisions under stationary controlled conditions.
Retention	Repeat the same instrumented task after a declared no-practice interval. The published three-month follow-up was a real-life self-report survey, not an objective retention test.
Transfer	Use a predeclared untrained object size, orientation or virtual route while preserving the same safety conditions.
Guardrail	Start with stationary soft targets. Never practise blindfolded walking near roads, stairs, water, heights or hazards; retain mobility aids and qualified orientation-and-mobility support.

Adult absolute-pitch naming

ID. EX-ABSOLUTE-PITCH

Narrow claim. Intensive perceptual training can improve pitch naming in adults, and a minority reach high all-pitch accuracy; success, generalisation and durability vary.

Table 33 — Capability Standard
Test	Definition
Direct	Accuracy across all twelve pitch classes on a predeclared instrument or timbre, scored over repeated balanced trials.
Retention	Repeat the balanced test after a declared no-practice interval of at least four weeks.
Transfer	Test an untrained timbre or octave declared before training.
Guardrail	Use comfortable listening levels and stop for pain, tinnitus, hearing change or distress. Never use medication or neurostimulation as a Titan training route.

Worked Exemplar: Visual Hyperacuity

Status: Worked exemplar — retained after primary-source verification as a non-module pathway. It does not create a seventeenth operational module and does not alter the Frozen Core.

Table 34 — Capability Standard
Field	Value
ID	EX-VISUAL-HYPERACUITY
Parent module	MOD-SENSORY-DISCRIMINATION
Evidence status	Verified Moderate — direct task learning; transfer limited and condition-dependent
Risk	Low / variable with visual symptoms
Acquisition mechanism	perceptual_calibration

Narrow claim

Practice can reduce thresholds for detecting tiny positional or orientation differences on defined visual tasks. It does not improve the optical focus of the eye, remove the need for corrective lenses, cure eye disease or produce telescopic vision.

Primary-source verification. Controlled studies show substantial practice-related improvement on Vernier or positional-acuity tasks in adults, but transfer is often narrow. Snell et al. trained seven adults on Vernier acuity for 7,500 trials and found improvement largely specific to the trained orientation; other studies show that transfer can occur under particular training designs. The evidence therefore supports task-specific threshold learning, not improved optics or broad visual enhancement. Evidence grade: Moderate for direct task learning; Limited for untrained transfer; long-term retention not established by the cited exemplar studies.

Assessment contract

Table 35 — Capability Standard
Element	Definition
Direct	Use a validated randomized alignment or orientation task and report threshold with viewing distance, display, correction, luminance, duration and retinal location.
Criterion	Predeclare the required threshold improvement and reliability rule across two sessions; do not select the best block after viewing results.
Retention	Repeat the identical threshold procedure after at least 28 days without practice.
Transfer	Test a predeclared untrained retinal location and a second untrained orientation, feature or stimulus configuration.
Burden	Record eyestrain, headache, migraine symptoms, dry eye, visual discomfort, concentration loss and session duration.

Protocol

Confirm that ordinary vision correction is current and comfortable.

Complete randomized baselines at trained and transfer locations.

Practise short blocks while rested; stop before accuracy deteriorates.

Keep display distance, scaling and ambient lighting consistent.

Use feedback during training but not declared assessments.

Preserve raw trial data and failed sessions.

Retest the original location after transfer training to detect interference.

Capability-maturity interpretation

Table 36 — Capability Standard
Field or stage	Application
functional	Threshold improves on the declared task.
transferred	Improvement appears at the predeclared novel location or feature.
automatic	Accuracy is preserved under a modest secondary task without unacceptable cost.
non_interfering	No deterioration in the original location, ordinary vision task or symptom status.
persistent	Improvement remains at delayed retest.

GUARDRAIL — Stop for pain, persistent blur, double vision, flashes, a new field defect, migraine escalation or other concerning visual change. Training is not a substitute for professional eye or neurological assessment.

Worked Exemplar: Olfactory Landmark Navigation

Status: Worked exemplar — retained after primary-source verification as a non-module pathway. It does not create a seventeenth operational module and does not alter the Frozen Core.

Table 37 — Capability Standard
Field	Value
ID	EX-OLFACTORY-LANDMARK-NAVIGATION
Parent modules	MOD-SENSORY-DISCRIMINATION; cross-link MOD-SPATIAL-SIMULATION
Evidence status	Verified Emerging–Moderate — controlled laboratory wayfinding only
Risk	Variable
Acquisition mechanism	latent_channel_recruitment; secondary perceptual_calibration

Narrow claim

People can learn stable odour landmarks and use them to support route learning in a controlled environment. This does not establish safe independent navigation, animal-like tracking ability, generally superior smell or diagnostic detection of disease or danger.

Primary-source verification. Hamburger and Knauff tested 24 participants in an odour-landmark wayfinding task and found above-chance use of 12 odours as orientation cues. Schwarz and Hamburger later tested 52 participants in a virtual maze and reported above-chance olfactory-landmark route decisions immediately and one month later, although performance was modest and the authors cautioned against broad ecological conclusions. Evidence grade: Emerging–Moderate for controlled laboratory wayfinding; Limited for transfer to ordinary real-world navigation.

Assessment contract

Table 38 — Capability Standard
Element	Definition
Direct	Complete a declared tabletop, seated virtual or cleared-room route using standardized odour landmarks; record identification accuracy, wrong turns, path efficiency, completion time and assistance.
Retention	Repeat a matched route after at least 28 days without olfactory-navigation practice.
Transfer	Use either the same odours in new positions or a matched new odour set in a new safe layout; predeclare which aspect is novel.
Burden	Record irritation, nausea, headache, sensory fatigue, aversion, asthma symptoms and setup burden.
Permitted route	Tabletop first; seated virtual second; cleared level indoor room only with a capable spotter and visible emergency fallback.

Protocol

Screen for asthma, allergy, migraine and fragrance sensitivity.

Use standardized low-irritant food-grade or professionally selected odour materials.

Establish identification accuracy before adding spatial-memory demand.

Train a small stable landmark set and change only one variable at a time.

Mask visible labels, container sounds and experimenter cues.

Include odour-free control trials where practical.

Stop immediately for respiratory, neurological or distress symptoms.

GUARDRAIL — Never use roads, public spaces, stairs, heights, open water, moving vehicles, machinery, unknown chemicals or deliberate irritant exposure. Never market this as medical scent detection or a replacement for smoke, carbon-monoxide or environmental sensors.

Worked Exemplar: Auditory Spatial Remapping

Status: Worked exemplar — retained after primary-source verification as a non-module pathway. It does not create a seventeenth operational module and does not alter the Frozen Core.

Table 39 — Capability Standard
Field	Value
ID	EX-AUDITORY-SPATIAL-REMAP
Parent module	MOD-SENSORY-DISCRIMINATION
Evidence status	Verified Moderate — adaptation to stable altered cues; broad transfer unproven
Risk	Variable
Acquisition mechanism	perceptual_remapping

Narrow claim

Adults can learn a deliberately altered relationship between spectral sound cues and spatial location under controlled conditions. This does not establish universally improved hearing, safer navigation or permanent remapping.

Primary-source verification. Hofman, Van Riswick and Van Opstal altered adult listeners’ pinna cues and found that elevation localisation initially deteriorated but was progressively relearned while ordinary localisation with the moulds removed remained available. The study supports adult recalibration to a stable altered cue mapping under controlled conditions. It does not establish permanent remapping, safety during movement or general hearing enhancement. Evidence grade: Moderate for direct adaptation; Limited for broad transfer and long-term retention.

Assessment contract

Table 40 — Capability Standard
Element	Definition
Direct	Identify randomized sound locations under one declared transformed mapping; record angular and elevation error, front-back reversals, response time and uncertainty.
Retention	Repeat after at least seven days and again at a later predeclared interval.
Transfer	Use untrained sound sources, frequencies and positions while preserving the mapping rule.
Native-function test	Repeat ordinary localization after training and at retention; any deterioration is interference.
Burden	Record auditory fatigue, headache, confusion, nausea, tinnitus change, disorientation and setup time.

Protocol

Complete baseline localization with ordinary audio.

Introduce one stable virtual spatial-audio transformation while stationary.

Begin with few positions and expand gradually.

Use trial-by-trial feedback during training, then remove it for assessment.

Test transfer with unfamiliar sounds and locations.

Retest ordinary localization to detect interference.

Schedule device-off retention and record relearning on return.

Capability-maturity interpretation

Table 41 — Capability Standard
Field or stage	Application
decoded	Above-chance localization using the transformed cue alone.
functional	Meets the direct angular-error and reversal criterion.
transferred	Meets criterion with unfamiliar sounds or positions.
incorporated	Combined-cue performance improves and weighting changes with cue reliability.
automatic	Performance survives a predeclared secondary task with acceptable cost.
non_interfering	Ordinary localization remains within declared tolerance.
persistent	Mapping remains available after a device-off interval.

GUARDRAIL — Training must be stationary. Do not walk, drive, cycle, cross roads, operate machinery or perform safety-critical work while hearing is experimentally transformed. Stop for new tinnitus, hearing change, vertigo, nausea, headache or distress.

Evidence links

EVD-ECHOLOCATION — historical evidence pointer only. Norman et al. (2021), doi:10.1371/journal.pone.0252330, remains source lineage and grants no current operational, publication or independent-navigation-safety authority.
EVD-PERCEPTUAL-LEARNING - Perceptual learning (Moderate)
EVD-OLFACTORY-TRAINING - Effects of olfactory training: a meta-analysis (Moderate)

Spatial Simulation & Cognitive Maps

Governing capability claim. Construct, update and use bounded spatial representations to plan routes, orient within a declared environment and predict spatial consequences.

Claim boundary. This record does not establish a universal sense of direction; safe navigation in hazardous environments; fitness to drive; perfect mental imagery.

Evidence contract. Direct: Route, pointing, layout-reconstruction or spatial-transformation accuracy on a declared environment or representation. Retention: Repeat the same map or route criterion after at least seven days without answer-bearing rehearsal. Transfer: Apply the mapping method to a novel but comparable layout, route or perspective. The claim remains task-, route-, support- and version-specific; structural conformance does not prove empirical validity.

HCE Genome panel · MOD-SPATIAL-SIMULATION · canonical registry record 1.0.0

HCE Genome panel · MOD-SPATIAL-SIMULATION · canonical registry record 1.0.0
HCE GENOME FIELD	CANONICAL MODULE CONFIGURATION
Identity & status	Spatial Simulation & Cognitive Maps · family Perception Spatial Cognition Sensorimotor Navigation · actor scope Individual · status Operational.
Purpose	Construct, update and use bounded spatial representations to plan routes, orient within a declared environment and predict spatial consequences.
Outcome & claim mode	Route, pointing, layout-reconstruction or spatial-transformation accuracy on a declared environment or representation. Claim mode: Confirmatory; formation route: Acquisition.
Claim boundary	Population: Adults or configured teams for whom the declared task and route are appropriate. Conditions: predeclared task; versioned protocol and criterion; declared support and constraints. Excludes: a universal sense of direction; safe navigation in hazardous environments; fitness to drive; perfect mental imagery.
Formation design	Primary mechanism: Procedural Compilation. Secondary: None declared. Patterns: calibration; scaffolding and fading; simulation and rehearsal.
Evidence & verification	Evidence grade: Strong; lifecycle: Not Assessed; mode: Instrumented. Baseline: Record performance before the formation protocol using the same or a matched scoring rule. Direct: Route, pointing, layout-reconstruction or spatial-transformation accuracy on a declared environment or representation. Retention: Repeat the same map or route criterion after at least seven days without answer-bearing rehearsal. Transfer: Apply the mapping method to a novel but comparable layout, route or perspective.
Failure & interference	Failure modes: route memorisation without map formation; landmark overdependence; map-environment mismatch; unsafe confidence. Interference plan: Measure and review: route memorisation without map formation, landmark overdependence, map-environment mismatch, unsafe confidence
Safety, consent & agency	Risk: Low. Voluntary participation with accessible withdrawal and no penalty Professional boundary: Use established mobility, navigation and local safety aids where consequences matter. Stop for disorientation, vertigo, distress or unsafe wayfinding. Agency boundary: Capability support must not erase consent, accountable ownership or an appropriate fallback route.
Dependencies & ecology	Prerequisites: basic perception or accessible representation; task instruction comprehension. Tools: None declared. Relationships: synergistic → MOD-SENSORY-DISCRIMINATION; enables → MOD-FORECASTING.
Economics & burden	Equipment: Task- and context-dependent; prefer the lowest-burden safe route. Opportunity cost: Formation, assessment, maintenance and context-switching burden. Maintenance cost: M2 — review every 30–90 days when operational use matters; remap after material environmental change.
Maintenance & retirement	Class M2; review due 90 days. Criterion: Repeat the declared direct criterion and review transfer, burden, safety and context change. Retirement triggers: criterion no longer met; unacceptable burden or safety event; invalid measure; loss of practical utility; plus 1 more.
Governance & provenance	Record HCE-TITAN-MOD-SPATIAL-001 · schema 0.2.0 · record version 1.0.0 · authority Titan · review 2026-10-24 · history append only with tombstones.

Table 43 — Capability Standard
Field	Value
ID	`MOD-SPATIAL-SIMULATION`
Status	new
Evidence	Strong
Risk	Low

Scope. Mental rotation, 2D-to-3D translation, landmark encoding, route preview, map reconstruction and mechanical visualisation.

Exclusions. Spatial-task improvement is not a claim of general-intelligence enhancement.

Measures

Table 44 — Capability Standard
Measure	Operational definition
Direct	Rotation, route, angle, landmark or reconstruction error on a defined task.
Retention	Repeat after at least seven days without rehearsal immediately beforehand.
Transfer	Complete a structurally related but untrained object, map or route task.
Burden	Visual strain, motion sensitivity, headache, vertigo or fatigue.

Protocol

Select one stationary spatial task.
Record time and error baseline.
Practise with feedback in short blocks.
Retest after a delay.
Run one predeclared novel-task test.
Use seated and non-VR variants when needed.

Level anchors

Table 45 — Capability Standard
Level	Module-specific anchor
L1	Baseline error or completion time is unstable on the selected task.
L2	Improves on repeated items with cues, examples or feedback.
L3	Meets the direct error and time criterion on unprompted trained-format items.
L4	Retains the gain and transfers it to a novel object, map or route task.
L5	Performs across defined formats or constraints and can teach the representation strategy accurately.

Guardrail. Use stationary, hazard-free tasks. Do not use experimental navigation drills around traffic or other real-world hazards.

Worked exemplars

Sketch-map reconstruction and controlled novel route

ID. EX-SKETCH-MAP-ROUTE

Narrow claim. Landmark, route and orientation representations can be practised and scored on defined maps and controlled routes; success does not establish safe independent navigation.

Table 46 — Capability Standard
Test	Definition
Direct	Study one fixed map for a predeclared interval, remove it, then reconstruct the landmarks, junctions, route order and orientation. Score landmark recall, placement error, route-sequence error and completion time against an answer key.
Retention	After at least seven days without using the original prompt, apply the same reconstruction method to a new matched map and repeat the identical scoring rules.
Transfer	Preview an unfamiliar, low-risk indoor or closed pedestrian route, then complete it without personal GPS prompts while a safety companion retains the route and may stop the trial. Score wrong turns, prompts and completion time.
Guardrail	Keep map work stationary. A route trial must be indoors or in a closed, hazard-free pedestrian setting with a safety companion; never use traffic, stairs, water, heights, wilderness or emergency routes, and never remove a mobility aid.

Evidence links

Governing capability claim. Coordinate perception and movement for a declared functional task while preserving safety, support and clinically appropriate boundaries.

Claim boundary. This record does not establish a toughness score; injury-proofing; fitness to drive; self-cleared rehabilitation; cold, pain or hypoxia tolerance as character.

Evidence contract. Direct: Task-specific balance, gait, reach, reaction, obstacle or movement-quality criterion using a declared instrument or qualified assessment. Retention: Repeat after a clinically appropriate delay without an answer-bearing warm-up. Transfer: Meet a comparable criterion in a predeclared safe variation of the task or environment. The claim remains task-, route-, support- and version-specific; structural conformance does not prove empirical validity.

Sensorimotor Intelligence & Fall Resilience

HCE Genome panel · MOD-SENSORIMOTOR-RESILIENCE · canonical registry record 1.0.0

HCE Genome panel · MOD-SENSORIMOTOR-RESILIENCE · canonical registry record 1.0.0
HCE GENOME FIELD	CANONICAL MODULE CONFIGURATION
Identity & status	Sensorimotor Intelligence & Fall Resilience · family Perception Spatial Cognition Sensorimotor Navigation · actor scope Individual · status Operational.
Purpose	Coordinate perception and movement for a declared functional task while preserving safety, support and clinically appropriate boundaries.
Outcome & claim mode	Task-specific balance, gait, reach, reaction, obstacle or movement-quality criterion using a declared instrument or qualified assessment. Claim mode: Confirmatory; formation route: Acquisition.
Claim boundary	Population: Adults or configured teams for whom the declared task and route are appropriate. Conditions: predeclared task; versioned protocol and criterion; declared support and constraints. Excludes: a toughness score; injury-proofing; fitness to drive; self-cleared rehabilitation; plus 1 more.
Formation design	Primary mechanism: Perceptual Calibration. Secondary: None declared. Patterns: calibration; graded exposure; scaffolding and fading.
Evidence & verification	Evidence grade: Moderate; lifecycle: Not Assessed; mode: Instrumented. Baseline: Record performance before the formation protocol using the same or a matched scoring rule. Direct: Task-specific balance, gait, reach, reaction, obstacle or movement-quality criterion using a declared instrument or qualified assessment. Retention: Repeat after a clinically appropriate delay without an answer-bearing warm-up. Transfer: Meet a comparable criterion in a predeclared safe variation of the task or environment.
Failure & interference	Failure modes: support removal; practice effect mistaken for function; unsafe progression; fatigue masking; non-equivalent task substitution. Interference plan: Measure and review: support removal, practice effect mistaken for function, unsafe progression, fatigue masking, non-equivalent task substitution
Safety, consent & agency	Risk: Professional Gate. Voluntary participation with accessible withdrawal and no penalty Professional boundary: Falls, neurological symptoms, acute injury, major dizziness and material mobility limitations require qualified assessment. Never remove required supports to make performance look stronger. Agency boundary: Capability support must not erase consent, accountable ownership or an appropriate fallback route.
Dependencies & ecology	Prerequisites: appropriate medical or professional clearance; safe environment and support. Tools: None declared. Relationships: synergistic → MOD-ASSISTIVE-FIT; synergistic → MOD-SPATIAL-SIMULATION.
Economics & burden	Equipment: Task- and context-dependent; prefer the lowest-burden safe route. Opportunity cost: Formation, assessment, maintenance and context-switching burden. Maintenance cost: M1–M2 according to consequence and clinical plan; reassess after falls, illness, injury or support change.
Maintenance & retirement	Class M1; review due 30 days. Criterion: Repeat the declared direct criterion and review transfer, burden, safety and context change. Retirement triggers: criterion no longer met; unacceptable burden or safety event; invalid measure; loss of practical utility; plus 1 more.
Governance & provenance	Record HCE-TITAN-MOD-SENSORIMOTOR-001 · schema 0.2.0 · record version 1.0.0 · authority Titan · review 2026-10-24 · history append only with tombstones.

Table 48 — Capability Standard
Field	Value
ID	`MOD-SENSORIMOTOR-RESILIENCE`
Status	new
Evidence	Moderate
Risk	Professional gate

Scope. Proprioceptive calibration, supported balance recovery, perception-to-action correction, motor imagery and approved cross-education.

Exclusions. Not a toughness challenge, fall-risk test or replacement for rehabilitation.

Measures

Table 49 — Capability Standard
Measure	Operational definition
Direct	Clinician-approved position-matching, movement-accuracy or supported recovery result.
Retention	Repeat the approved task after the specified clinical interval.
Transfer	Performance on a separately approved functional movement task.
Burden	Pain, dizziness, glucose disruption, near miss, assistance required or symptom change.

Protocol

Confirm professional clearance and physical support.
Select one approved low-risk task.
Record baseline accuracy and symptoms.
Practise at the prescribed dose.
Retest only within the approved plan.
Stop immediately for pain, dizziness, instability or new symptoms.

Level anchors

Table 50 — Capability Standard
Level	Module-specific anchor
L1	Approved baseline task requires substantial support or is inconsistent.
L2	Completes the approved task reliably with support and no adverse change.
L3	Meets the clinician-defined direct criterion under the approved conditions.
L4	Retains the gain and transfers it to a separately approved functional task.
L5	Maintains the approved performance across defined constraints and can describe the safety sequence accurately.

Guardrail. Balance, perturbation or rehabilitation work requires appropriate support and clinician approval when fall, fracture, metabolic or neurological risk is present.

Worked exemplars

Screen-based driving hazard perception

ID. EX-HAZARD-PERCEPTION

Narrow claim. Validated video tasks can measure latency to anticipate developing traffic hazards, and brief video-based training improved experienced drivers’ test performance with the benefit still evident after at least one week. This does not establish safer on-road driving, lower crash risk or fitness to drive.

Table 51 — Capability Standard
Test	Definition
Direct	On a one-exposure, expert-keyed set of unfamiliar licensed clips, report latency from developing-hazard onset to the first valid response, miss rate, false-response rate and invalid patterned responding. Training clips must never appear in the scored set.
Retention	After at least seven days, use a new matched clip set with the same onset-coding, latency and error rules.
Transfer	Use a predeclared unfamiliar screen-based set with a different road environment, weather, traffic density or jurisdiction. Transfer remains simulation-only and may not be tested through live driving, cycling or roadside exposure.
Guardrail	Use a stationary screen and licensed or owned clips. Never practise while controlling a vehicle or near live traffic. Stop for motion sickness, migraine, seizure warning, visual strain or trauma-related distress. A Titan result is not a driving licence, eyesight assessment, medical fitness decision or proof of safe driving.

Evidence links

EVD-PROPRIOCEPTION - The Effectiveness of Proprioceptive Training for Improving Motor Performance and Motor Dysfunction: A Systematic Review (Moderate)
EVD-BALANCE-PERTURBATION - Does perturbation-based balance training prevent falls? Systematic review and meta-analysis of preliminary randomized controlled trials (Moderate)

EVD-AOMI-2025 - Binks JA, Kenny RPW, van Schaik P, Wilson CJ, Lin D, Binks TJ, Eaves DL. The effects of combined action observation and motor imagery (AO+MI) practice on motor learning: a systematic review. Int Rev Sport Exerc Psychol. Published online 13 November 2025. DOI: 10.1080/1750984X.2025.2575511 (Moderate; 17 studies in healthy adults; database search completed March 2024; substantial heterogeneity in training schedules)

Combined action observation and motor imagery. Observing an action while simultaneously imagining performing it may be used as a supplementary practice route where physical practice is limited by injury, fatigue, access or safety. It is supplementary: the declared criterion remains physical performance, and retention and transfer must be tested by design rather than inferred from imagery adherence.

Governing capability claim. Acquire, retrieve and apply defined knowledge or skill after delay and in a predeclared novel context.

Claim boundary. This record does not establish photographic memory; unlimited memory; general-IQ increase; guaranteed far transfer; completion of training as proof of learning; fluency or familiarity as proof of retrieval.

Evidence contract. Direct: Uncued or minimally cued performance against a disclosed answer key or rubric after training support has been faded. Retention: Repeat after a delay appropriate to the use case, without relearning immediately before the test. Transfer: Apply the underlying rule, procedure or representation to novel material or a changed context declared before training. The claim remains task-, route-, support- and version-specific; structural conformance does not prove empirical validity.

Learning, Memory & Metacognition

Purpose. Acquire, retain, retrieve, transfer, generate and accurately monitor knowledge and skill.

Family boundary. No photographic memory, general-IQ increase, guaranteed genius or unlimited attention claim.

Learning, Memory & Skill Transfer

HCE Genome panel · MOD-LEARNING-TRANSFER · canonical registry record 1.0.0

HCE Genome panel · MOD-LEARNING-TRANSFER · canonical registry record 1.0.0
HCE GENOME FIELD	CANONICAL MODULE CONFIGURATION
Identity & status	Learning, Memory & Skill Transfer · family Learning Memory Creativity Metacognition · actor scope Individual · status Operational.
Purpose	Acquire, retrieve and apply defined knowledge or skill after delay and in a predeclared novel context.
Outcome & claim mode	Uncued or minimally cued performance against a disclosed answer key or rubric after training support has been faded. Claim mode: Confirmatory; formation route: Acquisition.
Claim boundary	Population: Adults or configured teams for whom the declared task and route are appropriate. Conditions: predeclared task; versioned protocol and criterion; declared support and constraints. Excludes: photographic memory; unlimited memory; general-IQ increase; guaranteed far transfer; plus 2 more.
Formation design	Primary mechanism: Procedural Compilation. Secondary: None declared. Patterns: retrieval and spacing; scaffolding and fading; interleaving; desirable difficulty; maintenance and reacquisition.
Evidence & verification	Evidence grade: Strong; lifecycle: Not Assessed; mode: Instrumented. Baseline: Record performance before the formation protocol using the same or a matched scoring rule. Direct: Uncued or minimally cued performance against a disclosed answer key or rubric after training support has been faded. Retention: Repeat after a delay appropriate to the use case, without relearning immediately before the test. Transfer: Apply the underlying rule, procedure or representation to novel material or a changed context declared before training.
Failure & interference	Failure modes: recognition mistaken for recall; overlearning one item set; testing too soon; transfer set leakage; prompt dependence; interleaving overload; retroactive interference; maintenance neglect. Interference plan: Measure and review: recognition mistaken for recall, overlearning one item set, testing too soon, transfer set leakage, prompt dependence
Safety, consent & agency	Risk: Low. Voluntary participation with accessible withdrawal and no penalty Professional boundary: Do not use learning protocols to delay assessment of cognitive decline, neurological symptoms, serious sleep problems or medication effects. Protect confidential learning records. Reduce load when practice drives marked distress, compulsive checking, sleep loss or functional deterioration. Agency boundary: Capability support must not erase consent, accountable ownership or an appropriate fallback route.
Dependencies & ecology	Prerequisites: basic access to material; task relevant prerequisites. Tools: None declared. Relationships: synergistic → MOD-METACOGNITIVE-CALIBRATION; enables → MOD-PROSPECTIVE-MEMORY; enables → MOD-CREATIVE-COGNITION.
Economics & burden	Equipment: Usually low equipment cost; time and attention are the principal costs. Opportunity cost: Acquisition time, feedback quality, interference and maintenance burden vary sharply by task. Maintenance cost: M1–M3 depending on decay and consequence: review windows must be task-specific rather than inferred from a universal memory curve.
Maintenance & retirement	Class M2; review due 90 days. Criterion: Repeat the declared direct criterion and review transfer, burden, safety and context change. Retirement triggers: criterion no longer met; unacceptable burden or safety event; invalid measure; loss of practical utility; plus 1 more.
Governance & provenance	Record HCE-TITAN-MOD-LEARNING-001 · schema 0.2.0 · record version 1.0.0 · authority Titan · review 2026-10-24 · history append only with tombstones.

Table 53 — Capability Standard
Field	Value
ID	`MOD-LEARNING-TRANSFER`
Status	promoted
Evidence	Strong
Risk	Low

Scope. Retrieval practice, spacing, feedback, retention and transfer for defined knowledge or skill.

Exclusions. No photographic-memory, unlimited-memory or broad-IQ claim.

Measures

Table 54 — Capability Standard
Measure	Operational definition
Direct	Uncued recall or task accuracy against a predeclared criterion.
Retention	Delayed recall or performance after a specified interval.
Transfer	Application to a new problem, example or context not used in practice.
Burden	Time, fatigue, frustration and interference with sleep or essential work.

Protocol

Define the material and final use.
Record an uncued baseline.
Schedule spaced retrieval with feedback.
Test after a delay.
Test one predeclared transfer task.
Adjust interval or method from misses.

Level anchors

Table 55 — Capability Standard
Level	Module-specific anchor
L1	Baseline recall or task accuracy is inconsistent without prompts.
L2	Meets criterion immediately after supported or repeated practice.
L3	Meets the direct criterion through uncued retrieval on two sessions.
L4	Retains the gain after the delay and succeeds on the transfer task.
L5	Retains and transfers across defined contexts and can teach the retrieval-and-feedback method.

Guardrail. Protect sleep and health. A memory strategy improves a defined task; it does not create unlimited capacity.

Worked exemplars

Task-specific motor learning

Governing capability claim. Estimate confidence, detect recurring error patterns and update strategy on bounded tasks without claiming general self-awareness.

Claim boundary. This record does not establish global self-knowledge; immunity to bias; diagnosis of cognitive disorder; certainty from confidence.

Evidence contract. Direct: Calibration gap, discrimination, confidence–accuracy relation or predeclared error-detection criterion on resolved items. Retention: Repeat calibration on a later matched item set after at least seven days. Transfer: Apply confidence tagging and error review to a different bounded task domain. The claim remains task-, route-, support- and version-specific; structural conformance does not prove empirical validity.

Table 56 — Capability Standard
Test	Definition
Direct	For a three-ball cascade, use the median successful catches across three sixty-second trials; accessible alternatives include an instrument phrase, knot or assembly sequence, seated hand skill or dumpling fold.
Retention	Repeat the identical scoring protocol after a declared no-practice interval.
Transfer	Use a predeclared alternate start, safe object set, phrase, material or closely related sequence.
Guardrail	Choose a safe accessible task and clear practice area. Do not use sharp, hot, heavy or breakable objects without appropriate competence and supervision.

Worked Exemplar: Calendar Engine

Status: Worked exemplar — retained after primary-source verification as a non-module pathway. It does not create a seventeenth operational module and does not alter the Frozen Core.

Table 57 — Capability Standard
Field	Value
ID	EX-CALENDAR-ENGINE
Parent module	MOD-LEARNING-TRANSFER
Cross-link	MOD-FORECASTING
Evidence status	Verified Moderate — bounded algorithmic acquisition; retention/far transfer limited
Risk	Low
Acquisition mechanism	procedural_compilation

Narrow claim

Explicit algorithmic training can enable adults to calculate the weekday of unfamiliar Gregorian dates with useful speed and accuracy. This does not establish general intelligence, spontaneous savant ability, exceptional autobiographical memory or broad numerical superiority.

Primary-source verification. Ward, Thompson-Lake, Ely and Kaminski trained sequence-space synaesthetes and non-synaesthete controls for approximately one hour across two weeks. In the third week, participants could select the weekday for dates drawn from several thousand possibilities in roughly ten seconds with about 80% accuracy; synaesthetes outperformed controls at final test. This supports rapid acquisition of a bounded algorithmic skill, not general intelligence or savant status. Evidence grade: Moderate for direct acquisition; Limited for month-scale retention and far transfer.

Why this belongs

Calendar calculation is a clean demonstration that an ability commonly perceived as mysterious can be acquired by compiling a bounded procedure. It is inexpensive, safe, objectively scored and suitable as an early proof-of-acquisition task.

Assessment contract

Table 58 — Capability Standard
Element	Definition
Direct	After training, answer at least 40 balanced, previously unseen Gregorian dates; record exact accuracy, median response time, omissions and error type.
Criterion	Predeclare accuracy and latency separately. A suggested pilot criterion is at least 80% correct with median response time no more than 15 seconds; this is not a universal scientific norm.
Retention	After at least 28 days without calendar-calculation practice, complete a new balanced set under the same conditions.
Transfer	Use an untrained century range and changed distribution of leap years, month positions and century anchors.
Burden	Record training minutes, frustration, headache, fatigue, strategy confusion and dropout.

Protocol

Complete 20 unseen baseline dates without instruction.

Learn century anchors, year offsets, month codes and leap-year correction.

Practise components separately before combining the procedure.

Use mixed retrieval rather than repeated blocks.

Reduce external prompts only after accuracy stabilizes.

Complete the declared direct test before viewing errors.

Schedule retention and transfer tests at declaration time.

Capability-maturity interpretation

Table 59 — Capability Standard
Field or stage	Application
L1	Cannot execute reliably; near chance or dependent on a complete worked example.
L2	Uses prompts or written scaffolding but remains slow or error-prone.
L3	Meets predeclared direct accuracy and latency in two sessions.
L4	Retains criterion and meets untrained-century transfer.
L5	Performs reliably across declared ranges and teaches the method without inflating the claim.

GUARDRAIL — Do not describe the result as savant syndrome, general intelligence enhancement, photographic memory or immunity to arithmetic error. Calendar software remains appropriate when operational accuracy matters.

Evidence links

EVD-RETRIEVAL-PRACTICE - Rethinking the Use of Tests: A Meta-Analysis of Practice Testing (Strong)
EVD-SPACING - Distributed practice in verbal recall tasks: A review and quantitative synthesis (Strong)
EVD-DELIBERATE-PRACTICE - Deliberate practice and performance in music, games, sports, education, and professions: a meta-analysis (Moderate)

EVD-ERROR-MANAGEMENT-2024 - Aliaga L, Bavolek RA, Cooper B, Mariorenzi A, Ahn J, Kraut A, Duong D, Burger C, Gisondi MA. Error Management Training and Adaptive Expertise in Learning Computed Tomography Interpretation: A Randomized Clinical Trial. JAMA Netw Open. 2024;7(9):e2431600. DOI: 10.1001/jamanetworkopen.2024.31600 (Moderate; 150 emergency medicine residents across seven residency programmes; three-arm trial)

EVD-SKILL-DECAY-2025 - Tatel CE, Ackerman PL. Procedural skill retention and decay: A meta-analytic review. Psychol Bull. 2025;151(6):696-736. DOI: 10.1037/bul0000481 (Strong; 1,344 effect sizes from 457 reports)

Adaptive-expertise note. Training under deliberately errorful and varied conditions is recorded as the acquisition mechanism error_management_acquisition. The claim is bounded: it concerns performance on novel cases within the trained domain. It does not establish general adaptability, resilience as a trait, or transfer to unrelated domains, and it carries a declared cost — routine-case accuracy and training time must both be measured, not assumed unchanged. Evidence boundary. The supporting trial studied 150 emergency medicine residents learning head computed tomography interpretation. That is a trained professional population learning one specialised cognitive task. Extension of the mechanism to general adult capability training is a plausible inference, not a demonstrated result, and must be declared as such wherever this acquisition mechanism is used.

Metacognitive Calibration & Error Detection

HCE Genome panel · MOD-METACOGNITIVE-CALIBRATION · canonical registry record 1.0.0

HCE Genome panel · MOD-METACOGNITIVE-CALIBRATION · canonical registry record 1.0.0
HCE GENOME FIELD	CANONICAL MODULE CONFIGURATION
Identity & status	Metacognitive Calibration & Error Detection · family Learning Memory Creativity Metacognition · actor scope Individual · status Operational.
Purpose	Estimate confidence, detect recurring error patterns and update strategy on bounded tasks without claiming general self-awareness.
Outcome & claim mode	Calibration gap, discrimination, confidence–accuracy relation or predeclared error-detection criterion on resolved items. Claim mode: Confirmatory; formation route: Acquisition.
Claim boundary	Population: Adults or configured teams for whom the declared task and route are appropriate. Conditions: predeclared task; versioned protocol and criterion; declared support and constraints. Excludes: global self-knowledge; immunity to bias; diagnosis of cognitive disorder; certainty from confidence.
Formation design	Primary mechanism: Procedural Compilation. Secondary: None declared. Patterns: calibration; prediction and resolution; error tagging.
Evidence & verification	Evidence grade: Moderate; lifecycle: Not Assessed; mode: Self Recorded. Baseline: Record performance before the formation protocol using the same or a matched scoring rule. Direct: Calibration gap, discrimination, confidence–accuracy relation or predeclared error-detection criterion on resolved items. Retention: Repeat calibration on a later matched item set after at least seven days. Transfer: Apply confidence tagging and error review to a different bounded task domain.
Failure & interference	Failure modes: confidence recorded after feedback; unresolved items scored as wrong; selective omission; global trait inference. Interference plan: Measure and review: confidence recorded after feedback, unresolved items scored as wrong, selective omission, global trait inference
Safety, consent & agency	Risk: Low. Voluntary participation with accessible withdrawal and no penalty Professional boundary: Do not turn confidence tracking into compulsive checking or a measure of personal worth. High-stakes decisions retain qualified review. Agency boundary: Capability support must not erase consent, accountable ownership or an appropriate fallback route.
Dependencies & ecology	Prerequisites: resolvable task feedback; basic recording access. Tools: None declared. Relationships: synergistic → MOD-LEARNING-TRANSFER; enables → MOD-FORECASTING; enables → MOD-HUMAN-AI-COMPLEMENTARITY.
Economics & burden	Equipment: Task- and context-dependent; prefer the lowest-burden safe route. Opportunity cost: Formation, assessment, maintenance and context-switching burden. Maintenance cost: M2 — periodic calibration windows; frequency increases where errors are consequential.
Maintenance & retirement	Class M2; review due 90 days. Criterion: Repeat the declared direct criterion and review transfer, burden, safety and context change. Retirement triggers: criterion no longer met; unacceptable burden or safety event; invalid measure; loss of practical utility; plus 1 more.
Governance & provenance	Record HCE-TITAN-MOD-METACOG-001 · schema 0.2.0 · record version 1.0.0 · authority Titan · review 2026-10-24 · history append only with tombstones.

Table 61 — Capability Standard
Field	Value
ID	`MOD-METACOGNITIVE-CALIBRATION`
Status	promoted
Evidence	Moderate
Risk	Low

Scope. Align confidence with actual performance, detect recurring error types and improve correction.

Exclusions. Not a global self-awareness or character score.

Measures

Table 62 — Capability Standard
Measure	Operational definition
Direct	Confidence–accuracy calibration and error-detection rate on a defined task.
Retention	Calibration on the same task family after a delay.
Transfer	Calibration on a related but untrained task.
Burden	Time, rumination, indecision or confidence suppression.

Protocol

Make an answer or decision before feedback.
Record confidence on a defined scale.
Compare confidence with the result.
Classify the error and correction.
Repeat after a delay.
Test calibration on a related task.

Governing capability claim. Generate, vary, select and test useful alternatives for a bounded problem under declared criteria.

Table 63 — Capability Standard
Level	Module-specific anchor
L1	Confidence and accuracy are poorly aligned or errors are rarely detected.
L2	Detects obvious errors after external feedback but calibration remains inconsistent.
L3	Meets the predeclared calibration or detection criterion on the trained task.
L4	Retains calibration and detects errors on a related untrained task.
L5	Maintains calibration across defined contexts and can explain recurring error patterns and corrections.

Claim boundary. This record does not establish general genius; permanent creative state; novelty without usefulness; insight as proof of truth.

Evidence contract. Direct: Number, diversity, novelty and usefulness of independently rated alternatives or prototypes on a declared brief. Retention: Repeat the generation-and-selection method after a delay without reusing prior answers. Transfer: Apply the method to a materially different problem class with a new rubric. The claim remains task-, route-, support- and version-specific; structural conformance does not prove empirical validity.

Worked exemplars

Confidence-tagged quizzing

ID. EX-CONFIDENCE-TAGGED-QUIZZING

Narrow claim. Recording confidence before feedback makes task-specific calibration and recurring error patterns measurable; it is not a general self-awareness score.

Table 64 — Capability Standard
Test	Definition
Direct	Answer a predeclared balanced set of at least twenty scorable items before feedback, attach a probability or fixed confidence bin to every answer, and calculate accuracy, calibration gap by bin and a proper score such as the Brier score when probabilities are used.
Retention	After at least seven days, repeat the same procedure on a new matched item set without the original correction notes visible.
Transfer	Use the same confidence scale and scoring method on a predeclared related but untrained topic or task format.
Guardrail	Use low-stakes material and a fixed review window. Do not use the exercise to diagnose a person, repeatedly reopen settled answers or replace qualified review of medical, legal, financial or safety-critical decisions.

Evidence links

EVD-METACOGNITION - Meta-analysis of Interventions for Monitoring Accuracy in Problem Solving (Moderate)

Creative Cognition & Insight

HCE Genome panel · MOD-CREATIVE-COGNITION · canonical registry record 1.0.0

HCE Genome panel · MOD-CREATIVE-COGNITION · canonical registry record 1.0.0
HCE GENOME FIELD	CANONICAL MODULE CONFIGURATION
Identity & status	Creative Cognition & Insight · family Learning Memory Creativity Metacognition · actor scope Individual · status Operational.
Purpose	Generate, vary, select and test useful alternatives for a bounded problem under declared criteria.
Outcome & claim mode	Number, diversity, novelty and usefulness of independently rated alternatives or prototypes on a declared brief. Claim mode: Confirmatory; formation route: Acquisition.
Claim boundary	Population: Adults or configured teams for whom the declared task and route are appropriate. Conditions: predeclared task; versioned protocol and criterion; declared support and constraints. Excludes: general genius; permanent creative state; novelty without usefulness; insight as proof of truth.
Formation design	Primary mechanism: Procedural Compilation. Secondary: None declared. Patterns: divergence then convergence; constraint switching; externalisation.
Evidence & verification	Evidence grade: Moderate; lifecycle: Not Assessed; mode: Self Recorded. Baseline: Record performance before the formation protocol using the same or a matched scoring rule. Direct: Number, diversity, novelty and usefulness of independently rated alternatives or prototypes on a declared brief. Retention: Repeat the generation-and-selection method after a delay without reusing prior answers. Transfer: Apply the method to a materially different problem class with a new rubric.
Failure & interference	Failure modes: idea count gaming; rater leakage; novelty bias; premature convergence; selection without testing. Interference plan: Measure and review: idea count gaming, rater leakage, novelty bias, premature convergence, selection without testing
Safety, consent & agency	Risk: Low. Voluntary participation with accessible withdrawal and no penalty Professional boundary: Separate generation from consequential selection. Medical, legal, financial and safety-critical ideas require qualified verification. Agency boundary: Capability support must not erase consent, accountable ownership or an appropriate fallback route.
Dependencies & ecology	Prerequisites: domain relevant inputs; declared evaluation rubric. Tools: None declared. Relationships: synergistic → MOD-LEARNING-TRANSFER; synergistic → MOD-ARGUMENTATION; enables → MOD-TOOL-REPAIR.
Economics & burden	Equipment: Task- and context-dependent; prefer the lowest-burden safe route. Opportunity cost: Formation, assessment, maintenance and context-switching burden. Maintenance cost: M2–M3; review when the method is needed rather than maintaining constant output.
Maintenance & retirement	Class M2; review due 90 days. Criterion: Repeat the declared direct criterion and review transfer, burden, safety and context change. Retirement triggers: criterion no longer met; unacceptable burden or safety event; invalid measure; loss of practical utility; plus 1 more.
Governance & provenance	Record HCE-TITAN-MOD-CREATIVE-001 · schema 0.2.0 · record version 1.0.0 · authority Titan · review 2026-10-24 · history append only with tombstones.

Table 66 — Capability Standard
Field	Value
ID	`MOD-CREATIVE-COGNITION`
Status	promoted
Evidence	Moderate
Risk	Low

Scope. Generate alternatives, select for novelty and usefulness, prototype, test and revise.

Exclusions. Training does not guarantee exceptional creative achievement or a general genius mode.

Measures

Table 67 — Capability Standard
Measure	Operational definition
Direct	Number and diversity of viable alternatives plus rated novelty and usefulness.
Retention	Repeat the method after a delay without the original prompts.
Transfer	Produce and test alternatives in a different problem domain.
Burden	Time, frustration, sleep loss or unfinished-work accumulation.

Protocol

Define the problem and usefulness criterion.
Separate generation from selection.
Generate genuinely different alternatives.
Select one using novelty plus usefulness.

Governing capability claim. Reliably execute a bounded future intention using proportionate internal or external cues.

Claim boundary. This record does not establish perfect memory; replacement for medication or emergency safeguards; alerts as proof of independence; unlimited commitments.

Evidence contract. Direct: Completion of a predeclared intention within its time or event window, with cue use and assistance recorded. Retention: Repeat the cueing method across a later observation window. Transfer: Apply the method to a different low-stakes intention type and observable cue. The claim remains task-, route-, support- and version-specific; structural conformance does not prove empirical validity.

Level anchors

Table 68 — Capability Standard
Level	Module-specific anchor
L1	Produces few distinct alternatives or selects without a criterion.
L2	Generates several alternatives with prompts or structured support.
L3	Independently generates, selects and tests an alternative against the criterion.
L4	Retains the process and transfers it to a different problem domain.
L5	Produces robust tested alternatives across defined constraints and can facilitate the process for another person.

Guardrail. Measure finished tests, not idea volume alone. Protect sleep and limit simultaneous unfinished prototypes.

Worked exemplars

Constraint-switch prototype sprint

ID. EX-CONSTRAINT-SWITCH-PROTOTYPE

Narrow claim. Separating generation from selection and testing makes the diversity, novelty and usefulness of alternatives observable on a bounded problem; it does not establish a general genius mode.

Table 69 — Capability Standard
Test	Definition
Direct	Define one low-risk problem and usefulness rubric, generate at least six alternatives spanning three genuinely different mechanisms, select one only after generation closes, then build a small test. Score distinct mechanisms, independent novelty and usefulness ratings, and whether the prototype met its criterion.
Retention	After at least seven days, repeat the complete process on a matched problem without the original prompts or alternatives visible.
Transfer	Apply the same generation-selection-test sequence to a predeclared problem in a different domain and preserve the same rubric structure.
Guardrail	Time-box generation and cap unfinished prototypes. Do not use the sprint for safety-critical design without qualified review, and stop if it displaces sleep or essential work.

Evidence links

EVD-CREATIVITY-TRAINING - The impact of creativity training on creative performance: A meta-analytic review and critical evaluation of 5 decades of creativity training studies (Moderate)

Prospective Memory & External Cues

HCE Genome panel · MOD-PROSPECTIVE-MEMORY · canonical registry record 1.0.0

HCE Genome panel · MOD-PROSPECTIVE-MEMORY · canonical registry record 1.0.0
HCE GENOME FIELD	CANONICAL MODULE CONFIGURATION
Identity & status	Prospective Memory & External Cues · family Learning Memory Creativity Metacognition · actor scope Individual · status Operational.
Purpose	Reliably execute a bounded future intention using proportionate internal or external cues.
Outcome & claim mode	Completion of a predeclared intention within its time or event window, with cue use and assistance recorded. Claim mode: Confirmatory; formation route: Acquisition.
Claim boundary	Population: Adults or configured teams for whom the declared task and route are appropriate. Conditions: predeclared task; versioned protocol and criterion; declared support and constraints. Excludes: perfect memory; replacement for medication or emergency safeguards; alerts as proof of independence; unlimited commitments.
Formation design	Primary mechanism: Support Environmental Redesign. Secondary: None declared. Patterns: implementation intention; cue anchoring; fallback preservation.
Evidence & verification	Evidence grade: Strong; lifecycle: Not Assessed; mode: Self Recorded. Baseline: Record performance before the formation protocol using the same or a matched scoring rule. Direct: Completion of a predeclared intention within its time or event window, with cue use and assistance recorded. Retention: Repeat the cueing method across a later observation window. Transfer: Apply the method to a different low-stakes intention type and observable cue.
Failure & interference	Failure modes: alert overload; ambiguous cue; unrecorded assistance; completion-window drift; compulsive checking. Interference plan: Measure and review: alert overload, ambiguous cue, unrecorded assistance, completion-window drift, compulsive checking
Safety, consent & agency	Risk: Low. Voluntary participation with accessible withdrawal and no penalty Professional boundary: Do not use this module as the sole safeguard for medication, childcare, driving, legal or emergency duties. Reduce commitments if cueing increases anxiety or compulsive checking. Agency boundary: Capability support must not erase consent, accountable ownership or an appropriate fallback route.
Dependencies & ecology	Prerequisites: observable event or time cue; accessible cue delivery. Tools: None declared. Relationships: synergistic → MOD-OPERATIONAL-SYSTEMS; synergistic → MOD-ASSISTIVE-FIT.
Economics & burden	Equipment: Task- and context-dependent; prefer the lowest-burden safe route. Opportunity cost: Formation, assessment, maintenance and context-switching burden. Maintenance cost: M1–M2 according to consequence; review cue fatigue and stale reminders weekly to monthly.
Maintenance & retirement	Class M1; review due 30 days. Criterion: Repeat the declared direct criterion and review transfer, burden, safety and context change. Retirement triggers: criterion no longer met; unacceptable burden or safety event; invalid measure; loss of practical utility; plus 1 more.
Governance & provenance	Record HCE-TITAN-MOD-PROSMEM-001 · schema 0.2.0 · record version 1.0.0 · authority Titan · review 2026-10-24 · history append only with tombstones.

Table 71 — Capability Standard
Field	Value
ID	`MOD-PROSPECTIVE-MEMORY`
Status	continued
Evidence	Strong
Risk	Low

Scope. Capture future intentions and attach them to reliable time- or event-based cues.

Exclusions. Evidence for reminders does not by itself establish a general capacity-budgeting system.

Measures

Table 72 — Capability Standard
Measure	Operational definition
Direct	Intention completion after the assigned cue.
Retention	Completion rate across at least two review periods.
Transfer	Reliable use of the cue method for a different intention type.
Burden	Alert count, ignored cues, anxiety and open-loop accumulation.

Protocol

Capture one intention outside your head.
Attach it to a time or observable event.
Define the smallest next action.
Set one reliable cue and fallback.
Review misses without blame.

Governing capability claim. Produce resolvable estimates, probabilities or quantities and improve calibration through scored outcomes.

Claim boundary. This record does not establish certainty; diagnosis; qualified legal, medical or financial advice; unscored intuition as forecasting.

Evidence contract. Direct: Brier score, calibration gap, interval coverage, Fermi estimate error or task-duration error on predeclared resolved items. Retention: Maintain calibration across a later rolling resolved-question window. Transfer: Apply the process to a new topic with comparable resolution quality. The claim remains task-, route-, support- and version-specific; structural conformance does not prove empirical validity.

Table 73 — Capability Standard
Level	Module-specific anchor
L1	Important intentions are often missed or depend on mental rehearsal.
L2	Completes cued intentions with repeated alerts or external support.
L3	Meets the completion criterion using one designed cue and fallback.
L4	Maintains completion across review periods and transfers the method to a different intention type.
L5	Runs a low-burden reliable cue system and can help another person design one without overload.

Guardrail. External memory should improve the declared task at acceptable burden. Disability, overload and executive constraints are design variables, not moral failures; an aid may be retained, changed or abandoned without penalty.

Worked exemplars

Cue-anchored daily task

ID. EX-CUE-ANCHORED-DAILY-TASK

Narrow claim. A future intention linked to one observable event cue and one fallback can be measured by timely completion and reminder burden; missed intentions are design data, not a moral score.

Table 74 — Capability Standard
Test	Definition
Direct	Choose one low-stakes daily task for fourteen days, link it to one observable event cue, define a completion window and one fallback, then record cue occurrence, completion within the window, fallback use and ignored-alert count.
Retention	Continue for a second fourteen-day period after removing any temporary coaching or redundant prompts while keeping the cue and fallback fixed.
Transfer	Apply the method to a different low-stakes intention type with a different observable cue, predeclaring the same completion and burden measures.
Guardrail	Do not use this exemplar as the sole safeguard for medication, childcare, driving, emergency, legal or other consequential duties. Reduce commitments before adding alerts, protect private logs and stop if reminders increase anxiety or compulsive checking.

Evidence links

EVD-PROSPECTIVE-REMINDERS - Strategic use of reminders: Influence of both domain-general and task-specific metacognitive confidence, independent of objective memory ability (Strong)

External-memory boundary. This module governs capture and cueing of future intentions, including where a wearable or assistant performs the capture. The capability is the reliable execution of the intention, not the possession of the device. Where capture is device-mediated, declare decay band D4 and retest unaided; a person whose intentions execute only while the device is running has a supported capability, which is a real result and must be recorded as supported rather than unaided.

Continuous personal capture — lifelogging, always-on audio or video, and passive multimodal recall aids — is not admitted to this module. It raises consent, third-party privacy and jurisdiction questions that the module's measures do not address, and it is held in the Research Observatory rather than tested here.

Quantitative, Decision & Causal Intelligence

Purpose. Estimate uncertainty, choose actions, test causal explanations and learn from resolved outcomes.

Family boundary. No immunity to bias, certainty from a probability or replacement for qualified professional advice.

Quantitative Intelligence & Forecasting

HCE Genome panel · MOD-FORECASTING · canonical registry record 1.0.0

HCE Genome panel · MOD-FORECASTING · canonical registry record 1.0.0
HCE GENOME FIELD	CANONICAL MODULE CONFIGURATION
Identity & status	Quantitative Intelligence & Forecasting · family Quantitative Decision Causal Intelligence · actor scope Individual · status Operational.
Purpose	Produce resolvable estimates, probabilities or quantities and improve calibration through scored outcomes.
Outcome & claim mode	Brier score, calibration gap, interval coverage, Fermi estimate error or task-duration error on predeclared resolved items. Claim mode: Confirmatory; formation route: Acquisition.
Claim boundary	Population: Adults or configured teams for whom the declared task and route are appropriate. Conditions: predeclared task; versioned protocol and criterion; declared support and constraints. Excludes: certainty; diagnosis; qualified legal, medical or financial advice; unscored intuition as forecasting.
Formation design	Primary mechanism: Procedural Compilation. Secondary: None declared. Patterns: base rate first; decomposition; calibration; update on evidence.
Evidence & verification	Evidence grade: Strong; lifecycle: Not Assessed; mode: Instrumented. Baseline: Record performance before the formation protocol using the same or a matched scoring rule. Direct: Brier score, calibration gap, interval coverage, Fermi estimate error or task-duration error on predeclared resolved items. Retention: Maintain calibration across a later rolling resolved-question window. Transfer: Apply the process to a new topic with comparable resolution quality.
Failure & interference	Failure modes: hindsight editing; vague resolution; base-rate neglect; false precision; selective scoring. Interference plan: Measure and review: hindsight editing, vague resolution, base-rate neglect, false precision, selective scoring
Safety, consent & agency	Risk: Low. Voluntary participation with accessible withdrawal and no penalty Professional boundary: Forecasts inform decisions but never erase uncertainty or replace qualified advice. Define invalid outcomes before resolution. Agency boundary: Capability support must not erase consent, accountable ownership or an appropriate fallback route.
Dependencies & ecology	Prerequisites: resolvable questions; reliable resolution source. Tools: None declared. Relationships: enables → MOD-CAUSAL-EXPERIMENTAL; synergistic → MOD-METACOGNITIVE-CALIBRATION; enables → MOD-TEAM-ORCHESTRATION.
Economics & burden	Equipment: Task- and context-dependent; prefer the lowest-burden safe route. Opportunity cost: Formation, assessment, maintenance and context-switching burden. Maintenance cost: M1–M2 using rolling resolved windows; archive stale unresolved questions.
Maintenance & retirement	Class M1; review due 30 days. Criterion: Repeat the declared direct criterion and review transfer, burden, safety and context change. Retirement triggers: criterion no longer met; unacceptable burden or safety event; invalid measure; loss of practical utility; plus 1 more.
Governance & provenance	Record HCE-TITAN-MOD-FORECAST-001 · schema 0.2.0 · record version 1.0.0 · authority Titan · review 2026-10-24 · history append only with tombstones.

Table 76 — Capability Standard
Field	Value
ID	`MOD-FORECASTING`
Status	continued
Evidence	Strong
Risk	Low

Scope. Resolvable probability estimates, base rates, decomposition, calibration and evidence-based updating.

Exclusions. A forecast is not certainty, diagnosis or qualified legal, medical or financial advice.

Measures

Table 77 — Capability Standard
Measure	Operational definition
Direct	Brier score, calibration gap or Fermi estimate error.
Retention	Calibration across a rolling resolved-question window.
Transfer	Calibration on a new topic with comparable resolution quality.
Burden	Research time, unresolved-question load and false precision.

Protocol

Governing capability claim. Choose actions and test bounded causal explanations using predeclared comparisons, outcomes and stopping rules.

Claim boundary. This record does not establish proof from correlation; uncontrolled self-experimentation; medical treatment selection without clinicians; certainty from one trial.

Evidence contract. Direct: Decision quality or causal estimate against a declared outcome, comparison rule and eligibility set. Retention: Repeat the reasoning and documentation method on a later decision or experiment. Transfer: Apply it to a different low-risk domain with a new outcome and confound map. The claim remains task-, route-, support- and version-specific; structural conformance does not prove empirical validity.

Find a relevant base rate.
Publish a probability.
Update only for relevant evidence.
Resolve and score.

Level anchors

Table 78 — Capability Standard
Level	Module-specific anchor
L1	Uses vague confidence language or cannot reliably resolve and score forecasts.
L2	Records probabilities and outcomes with support but calibration is unstable.
L3	Maintains a resolvable forecast set and meets the predeclared rolling calibration criterion.
L4	Retains calibration and transfers the process to a new topic.
L5	Shows robust calibration across defined topics and can teach resolution, base-rate and update discipline.

Guardrail. Forecasts inform decisions; they never erase uncertainty or replace qualified advice.

Worked exemplars

Task-duration forecasting

ID. EX-TASK-DURATION

Narrow claim. Repeated estimates can be calibrated against actual task duration; the direction and size of bias must be measured rather than assumed.

Table 79 — Capability Standard
Test	Definition
Direct	For at least ten comparable tasks, record the estimate and actual active time, then calculate signed error and absolute percentage or log-ratio error.
Retention	Repeat the method across a later rolling window without hiding overruns or abandoned tasks.
Transfer	Predeclare a different task class and test whether an outside-view estimate based on completed work remains calibrated.
Guardrail	Separate active time from waiting time, keep invalidation rules fixed and never turn a forecast into a promise of certainty.

Evidence links

EVD-FORECASTING - Psychological Strategies for Winning a Geopolitical Forecasting Tournament (Strong)

Decision, Causal & Experimental Intelligence

HCE Genome panel · MOD-CAUSAL-EXPERIMENTAL · canonical registry record 1.0.0

HCE Genome panel · MOD-CAUSAL-EXPERIMENTAL · canonical registry record 1.0.0
HCE GENOME FIELD	CANONICAL MODULE CONFIGURATION
Identity & status	Decision, Causal & Experimental Intelligence · family Quantitative Decision Causal Intelligence · actor scope Individual · status Operational.
Purpose	Choose actions and test bounded causal explanations using predeclared comparisons, outcomes and stopping rules.
Outcome & claim mode	Decision quality or causal estimate against a declared outcome, comparison rule and eligibility set. Claim mode: Confirmatory; formation route: Acquisition.
Claim boundary	Population: Adults or configured teams for whom the declared task and route are appropriate. Conditions: predeclared task; versioned protocol and criterion; declared support and constraints. Excludes: proof from correlation; uncontrolled self-experimentation; medical treatment selection without clinicians; certainty from one trial.
Formation design	Primary mechanism: Procedural Compilation. Secondary: None declared. Patterns: decision journal; n of 1; counterfactual testing; stop rule.
Evidence & verification	Evidence grade: Moderate; lifecycle: Not Assessed; mode: Self Recorded. Baseline: Record performance before the formation protocol using the same or a matched scoring rule. Direct: Decision quality or causal estimate against a declared outcome, comparison rule and eligibility set. Retention: Repeat the reasoning and documentation method on a later decision or experiment. Transfer: Apply it to a different low-risk domain with a new outcome and confound map.
Failure & interference	Failure modes: outcome switching; confounding; optional stopping; regression to mean; single-case overgeneralisation. Interference plan: Measure and review: outcome switching, confounding, optional stopping, regression to mean, single-case overgeneralisation
Safety, consent & agency	Risk: Variable. Voluntary participation with accessible withdrawal and no penalty Professional boundary: Material-risk health, medication, legal and financial experiments require qualified governance. Do not manipulate others without consent. Agency boundary: Capability support must not erase consent, accountable ownership or an appropriate fallback route.
Dependencies & ecology	Prerequisites: defined outcome; ethical and safe comparison. Tools: None declared. Relationships: synergistic → MOD-FORECASTING; synergistic → MOD-ARGUMENTATION; enables → MOD-OPERATIONAL-SYSTEMS.
Economics & burden	Equipment: Task- and context-dependent; prefer the lowest-burden safe route. Opportunity cost: Formation, assessment, maintenance and context-switching burden. Maintenance cost: M2; review decision journals after outcomes resolve and retire experiments that do not earn their burden.
Maintenance & retirement	Class M2; review due 90 days. Criterion: Repeat the declared direct criterion and review transfer, burden, safety and context change. Retirement triggers: criterion no longer met; unacceptable burden or safety event; invalid measure; loss of practical utility; plus 1 more.
Governance & provenance	Record HCE-TITAN-MOD-CAUSAL-001 · schema 0.2.0 · record version 1.0.0 · authority Titan · review 2026-10-24 · history append only with tombstones.

Table 81 — Capability Standard
Field	Value
ID	`MOD-CAUSAL-EXPERIMENTAL`
Status	new
Evidence	Moderate
Risk	Variable

Scope. Causal diagrams, alternative explanations, confounders, expected value, value of information and safe experiment design.

Exclusions. No general immunity to bias or permission for unsafe self-experimentation.

Measures

Table 82 — Capability Standard
Measure	Operational definition
Direct	Causal-structure error, confounders identified, decision-criterion fit or experiment-validity checklist.
Retention	Repeat the analysis after a delay without viewing the original answer.
Transfer	Apply the method to a different decision or causal problem.
Burden	Decision delay, analysis cost, experiment risk and opportunity cost.

Protocol

State the decision and objective.
Draw the assumed causal path.

Governing capability claim. Construct, test and revise bounded claims while preserving evidence limits, uncertainty and good-faith disagreement.

Claim boundary. This record does not establish winning as truth; general lie detection; rhetorical fluency as evidence; certainty from debate.

Evidence contract. Direct: Rubric-scored claim structure, source quality, counterargument handling and justified update on a declared issue. Retention: Repeat the method after delay on a matched issue without answer-bearing notes. Transfer: Apply it to a new domain with different sources and incentives. The claim remains task-, route-, support- and version-specific; structural conformance does not prove empirical validity.

Predeclare outcome, stop rule and interpretation limits.
Review prediction versus result.

Level anchors

Table 83 — Capability Standard
Level	Module-specific anchor
L1	Causal claims omit major alternatives or the decision lacks an explicit objective.
L2	Identifies obvious confounders and criteria with prompts or templates.
L3	Completes the causal and decision checklist and meets the direct-validity criterion independently.
L4	Retains the method and applies it successfully to a different decision or experiment.
L5	Uses the method robustly across defined decisions and can review another analysis while preserving uncertainty and safety.

Guardrail. Do not run health, substance, financial, legal or high-risk experiments without appropriate professional oversight. Prefer reversible, low-risk tests.

Worked exemplars

Low-risk N-of-1 time-of-day experiment

ID. EX-NOF1-TIME-OF-DAY

Narrow claim. A randomized or counterbalanced within-person comparison can test whether performance on one safe task differs between two feasible time windows; it does not identify a universal chronotype or prove a medical cause.

Table 84 — Capability Standard
Test	Definition
Direct	Predeclare one low-risk repeatable task, one primary outcome, two feasible time windows, exclusion rules and a stop rule. Randomize or counterbalance at least twelve valid sessions while holding task format and scoring constant, then report each condition, missingness, the within-person difference and uncertainty without deleting inconvenient sessions.
Retention	After at least seven days, independently design and run a shorter confirmatory block with the same primary outcome and fixed interpretation rule.
Transfer	Apply the design checklist to a different reversible low-risk task while predeclaring a new outcome and likely confounders.
Guardrail	Never manipulate sleep duration, medication, substances, meals needed for health, driving or hazardous work. Do not infer a diagnosis or reorganise consequential duties from a small personal result; stop for adverse symptoms or material life disruption.

Evidence links

EVD-CAUSAL-TRAINING - Causal theory error in college students’ understanding of science studies (Moderate)
EVD-DEBIASING - Debiasing Decisions: Improved Decision Making With a Single Training Intervention (Moderate)

Communication, Argument & Negotiation

Purpose. Listen, explain, teach, test claims, negotiate and repair understanding without sacrificing truth or consent.

Family boundary. No coercion, manipulation, harassment, deception or exploitation of vulnerability.

Argumentation & Dialectical Testing

HCE Genome panel · MOD-ARGUMENTATION · canonical registry record 1.0.0

HCE Genome panel · MOD-ARGUMENTATION · canonical registry record 1.0.0
HCE GENOME FIELD	CANONICAL MODULE CONFIGURATION
Identity & status	Argumentation & Dialectical Testing · family Communication Argument Negotiation · actor scope Individual · status Operational.
Purpose	Construct, test and revise bounded claims while preserving evidence limits, uncertainty and good-faith disagreement.
Outcome & claim mode	Rubric-scored claim structure, source quality, counterargument handling and justified update on a declared issue. Claim mode: Confirmatory; formation route: Acquisition.
Claim boundary	Population: Adults or configured teams for whom the declared task and route are appropriate. Conditions: predeclared task; versioned protocol and criterion; declared support and constraints. Excludes: winning as truth; general lie detection; rhetorical fluency as evidence; certainty from debate.
Formation design	Primary mechanism: Procedural Compilation. Secondary: None declared. Patterns: steelman; claim evidence warrant; disconfirmation search.
Evidence & verification	Evidence grade: Moderate; lifecycle: Not Assessed; mode: Self Recorded. Baseline: Record performance before the formation protocol using the same or a matched scoring rule. Direct: Rubric-scored claim structure, source quality, counterargument handling and justified update on a declared issue. Retention: Repeat the method after delay on a matched issue without answer-bearing notes. Transfer: Apply it to a new domain with different sources and incentives.
Failure & interference	Failure modes: strawmanning; source laundering; goalpost movement; identity defence; argument from fluency. Interference plan: Measure and review: strawmanning, source laundering, goalpost movement, identity defence, argument from fluency
Safety, consent & agency	Risk: Low. Voluntary participation with accessible withdrawal and no penalty Professional boundary: Do not use coercion, harassment or deceptive framing. Consequential claims require authoritative sources and qualified review. Agency boundary: Capability support must not erase consent, accountable ownership or an appropriate fallback route.
Dependencies & ecology	Prerequisites: source access; basic comprehension. Tools: None declared. Relationships: synergistic → MOD-TEACHING-NEGOTIATION; synergistic → MOD-CAUSAL-EXPERIMENTAL; enables → MOD-HUMAN-AI-COMPLEMENTARITY.
Economics & burden	Equipment: Task- and context-dependent; prefer the lowest-burden safe route. Opportunity cost: Formation, assessment, maintenance and context-switching burden. Maintenance cost: M2; periodic adversarial review of consequential claims.
Maintenance & retirement	Class M2; review due 90 days. Criterion: Repeat the declared direct criterion and review transfer, burden, safety and context change. Retirement triggers: criterion no longer met; unacceptable burden or safety event; invalid measure; loss of practical utility; plus 1 more.
Governance & provenance	Record HCE-TITAN-MOD-ARGUMENT-001 · schema 0.2.0 · record version 1.0.0 · authority Titan · review 2026-10-24 · history append only with tombstones.

Table 86 — Capability Standard
Field	Value
ID	`MOD-ARGUMENTATION`
Status	continued
Evidence	Moderate
Risk	Low

Governing capability claim. Create usable mutual understanding, teach a bounded concept and negotiate acceptable agreements without sacrificing truth or consent.

Claim boundary. This record does not establish agreement as understanding; coercive persuasion; universal empathy; guaranteed relationship repair.

Evidence contract. Direct: Learner reconstruction, delayed application, listening accuracy or negotiated-outcome rubric on a declared interaction. Retention: Repeat the method after delay with the same or a matched participant and no answer-bearing prompt. Transfer: Apply it to a different topic, learner or low-stakes negotiation context. The claim remains task-, route-, support- and version-specific; structural conformance does not prove empirical validity.

Measures

Table 87 — Capability Standard
Measure	Operational definition
Direct	Claim–evidence fit, counterargument fidelity and agreed crux.
Retention	Repeat evaluation of a new claim after a delay.
Transfer	Apply the method in a different topic or communication format.
Burden	Conflict, defensiveness, time and relationship cost.

Protocol

State one falsifiable claim.
Separate source, observation and assumption.
Write the strongest opposing case.
Name the crux.
State what changes your mind.
Update confidence after evidence.

Level anchors

Table 88 — Capability Standard
Level	Module-specific anchor
L1	Claims, evidence and assumptions are mixed or opposition is caricatured.
L2	Separates components and improves the counterargument with prompts.
L3	Produces a faithful argument map and identifies a testable crux independently.
L4	Retains the method and transfers it to a different topic or format.
L5	Facilitates rigorous disagreement across defined contexts while preserving truth, consent and relationship safety.

Guardrail. Persuasion must not become coercion, deception, harassment or exploitation.

Worked exemplars

Claim-and-counterargument decision memo

ID. EX-DECISION-MEMO

Narrow claim. A short falsifiable memo can make evidence, assumptions, strongest opposition, cruxes and confidence updates auditable.

Table 89 — Capability Standard
Test	Definition
Direct	Use a fixed rubric for source-claim fit, opposing-case fidelity, crux quality and justified confidence updating.
Retention	Repeat the method after a delay without the original checklist visible.
Transfer	Apply the same rubric to a different low-stakes topic with new sources.
Guardrail	Do not use argument skill to harass, coerce or overwhelm. High-stakes medical, legal and financial claims still require qualified expertise.

Evidence links

EVD-DIALECTICAL-ESTIMATION - The Wisdom of Many in One Mind: Improving Individual Judgments With Dialectical Bootstrapping (Moderate)

Listening, Teaching & Negotiation

HCE Genome panel · MOD-TEACHING-NEGOTIATION · canonical registry record 1.0.0

HCE Genome panel · MOD-TEACHING-NEGOTIATION · canonical registry record 1.0.0
HCE GENOME FIELD	CANONICAL MODULE CONFIGURATION
Identity & status	Listening, Teaching & Negotiation · family Communication Argument Negotiation · actor scope Individual · status Operational.
Purpose	Create usable mutual understanding, teach a bounded concept and negotiate acceptable agreements without sacrificing truth or consent.
Outcome & claim mode	Learner reconstruction, delayed application, listening accuracy or negotiated-outcome rubric on a declared interaction. Claim mode: Confirmatory; formation route: Acquisition.
Claim boundary	Population: Adults or configured teams for whom the declared task and route are appropriate. Conditions: predeclared task; versioned protocol and criterion; declared support and constraints. Excludes: agreement as understanding; coercive persuasion; universal empathy; guaranteed relationship repair.
Formation design	Primary mechanism: Procedural Compilation. Secondary: None declared. Patterns: teach back; active listening; interest mapping; consent preservation.
Evidence & verification	Evidence grade: Moderate; lifecycle: Not Assessed; mode: Self Recorded. Baseline: Record performance before the formation protocol using the same or a matched scoring rule. Direct: Learner reconstruction, delayed application, listening accuracy or negotiated-outcome rubric on a declared interaction. Retention: Repeat the method after delay with the same or a matched participant and no answer-bearing prompt. Transfer: Apply it to a different topic, learner or low-stakes negotiation context.
Failure & interference	Failure modes: fluency mistaken for learning; leading teach-back; hidden coercion; agreement without implementation; power imbalance ignored. Interference plan: Measure and review: fluency mistaken for learning, leading teach-back, hidden coercion, agreement without implementation, power imbalance ignored
Safety, consent & agency	Risk: Low. Voluntary participation with accessible withdrawal and no penalty Professional boundary: Preserve consent, withdrawal, confidentiality and power-awareness. Abuse, safeguarding and legal disputes require appropriate professional routes. Agency boundary: Capability support must not erase consent, accountable ownership or an appropriate fallback route.
Dependencies & ecology	Prerequisites: shared communication channel; declared interaction goal. Tools: None declared. Relationships: synergistic → MOD-ARGUMENTATION; enables → MOD-TEAM-ORCHESTRATION.
Economics & burden	Equipment: Task- and context-dependent; prefer the lowest-burden safe route. Opportunity cost: Formation, assessment, maintenance and context-switching burden. Maintenance cost: M2; review after consequential interactions or role changes.
Maintenance & retirement	Class M2; review due 90 days. Criterion: Repeat the declared direct criterion and review transfer, burden, safety and context change. Retirement triggers: criterion no longer met; unacceptable burden or safety event; invalid measure; loss of practical utility; plus 1 more.
Governance & provenance	Record HCE-TITAN-MOD-TEACHNEG-001 · schema 0.2.0 · record version 1.0.0 · authority Titan · review 2026-10-24 · history append only with tombstones.

Table 91 — Capability Standard
Field	Value
ID	`MOD-TEACHING-NEGOTIATION`
Status	promoted
Evidence	Moderate
Risk	Low

Scope. Listening accuracy, explanation, teach-back, negotiation, consent and relationship repair.

Exclusions. No guarantee of agreement and no permission to manipulate or pressure.

Measures

Table 92 — Capability Standard
Measure	Operational definition
Direct	Listener comprehension, teach-back accuracy, listening reconstruction or negotiated criterion.
Retention	Learner performance or agreement quality after a delay.
Transfer	Successful explanation or negotiation with a different person or topic.
Burden	Time, distress, power imbalance and relationship cost.

Governing capability claim. Combine independent information, expertise, dissent and coordination into better group performance on a declared task.

Claim boundary. This record does not establish group harmony as intelligence; individual personality scoring; majority vote as truth; unaccountable diffusion of responsibility.

Evidence contract. Direct: Team outcome against a disclosed rubric plus independence, participation, error and coordination measures. Retention: Repeat the operating method with the same team after delay or across a later cycle. Transfer: Apply the orchestration method to a new team task with changed roles or information distribution. The claim remains task-, route-, support- and version-specific; structural conformance does not prove empirical validity.

Listen and reconstruct before responding.
Explain in the listener’s terms.
Use teach-back or an explicit agreement check.
Record unresolved differences.
Review outcome and relationship cost.

Level anchors

Table 93 — Capability Standard
Level	Module-specific anchor
L1	Understanding or agreement frequently fails and is not checked.
L2	Uses reconstruction or teach-back with prompts and repairs obvious misunderstandings.
L3	Meets the comprehension or negotiated-outcome criterion on the defined interaction.
L4	Retains the process and transfers it to a different person or topic.
L5	Performs across defined communication constraints and can teach the consent, listening and repair sequence.

Guardrail. Account for power differences. Stop if an interaction becomes coercive, unsafe or beyond your competence to mediate.

Worked exemplars

Teach-back with a delayed learner test

ID. EX-TEACH-BACK-DELAYED-LEARNER

Narrow claim. A learner’s immediate reconstruction and delayed application can test whether one bounded explanation produced usable understanding; fluency or agreement alone is not evidence of learning.

Table 94 — Capability Standard
Test	Definition
Direct	With consent, teach one bounded low-stakes concept, then have the learner explain it in their own words and solve one near application without copying the explanation. Score a predeclared accuracy and misconception rubric.
Retention	After at least seven days without reteaching, have the learner complete a new matched explanation or application and score it with the same rubric.
Transfer	Use the same consent, reconstruction and delayed-test sequence with a different learner or topic and a new answer key.
Guardrail	Keep the exercise consensual and account for power, language, disability and prior knowledge. Do not test high-stakes clinical, legal, financial or safety instruction unless a qualified professional owns the content and assessment.

Evidence qualification. Teach-back has shown mixed but generally positive effects on knowledge, self-management and related outcomes in heterogeneous health-education studies. This exemplar therefore supports a bounded comprehension check, while any claim about delayed retention or transfer must be tested directly. (R7D-S19)

Evidence links

EVD-COMMUNICATION-TRAINING - Communication skills training for healthcare professionals working with people who have cancer (Moderate)
EVD-ASSERTIVENESS - The effectiveness of assertiveness communication training programs for healthcare professionals and students: A systematic review (Moderate)

Collaboration & Collective Intelligence

Purpose. Combine independent information, expertise, dissent, coordination and shared memory into better group performance.

Family boundary. A group property must not be presented as an individual personality score or inferred from identity.

Collective Intelligence & Team Orchestration

HCE Genome panel · MOD-TEAM-ORCHESTRATION · canonical registry record 1.0.0

HCE Genome panel · MOD-TEAM-ORCHESTRATION · canonical registry record 1.0.0
HCE GENOME FIELD	CANONICAL MODULE CONFIGURATION
Identity & status	Collective Intelligence & Team Orchestration · family Collaboration Collective Intelligence · actor scope Team · status Operational.
Purpose	Combine independent information, expertise, dissent and coordination into better group performance on a declared task.
Outcome & claim mode	Team outcome against a disclosed rubric plus independence, participation, error and coordination measures. Claim mode: Confirmatory; formation route: Acquisition.
Claim boundary	Population: Adults or configured teams for whom the declared task and route are appropriate. Conditions: predeclared task; versioned protocol and criterion; declared support and constraints. Excludes: group harmony as intelligence; individual personality scoring; majority vote as truth; unaccountable diffusion of responsibility.
Formation design	Primary mechanism: Collective Emergence. Secondary: None declared. Patterns: independent estimates first; role clarity; dissent preservation; shared memory.
Evidence & verification	Evidence grade: Strong; lifecycle: Not Assessed; mode: Self Recorded. Baseline: Record performance before the formation protocol using the same or a matched scoring rule. Direct: Team outcome against a disclosed rubric plus independence, participation, error and coordination measures. Retention: Repeat the operating method with the same team after delay or across a later cycle. Transfer: Apply the orchestration method to a new team task with changed roles or information distribution.
Failure & interference	Failure modes: premature convergence; social loafing; dominance effects; information silos; diffused accountability. Interference plan: Measure and review: premature convergence, social loafing, dominance effects, information silos, diffused accountability
Safety, consent & agency	Risk: Low. Voluntary participation with accessible withdrawal and no penalty Professional boundary: Protect dissent, confidentiality, consent and accountable ownership. Do not infer individual worth from a group result. Agency boundary: Capability support must not erase consent, accountable ownership or an appropriate fallback route.
Dependencies & ecology	Prerequisites: multiple contributors; shared task and governance. Tools: None declared. Relationships: synergistic → MOD-FORECASTING; synergistic → MOD-TEACHING-NEGOTIATION; enables → MOD-OPERATIONAL-SYSTEMS.
Economics & burden	Equipment: Task- and context-dependent; prefer the lowest-burden safe route. Opportunity cost: Formation, assessment, maintenance and context-switching burden. Maintenance cost: M1–M2 through recurring retrospectives and role review.
Maintenance & retirement	Class M1; review due 30 days. Criterion: Repeat the declared direct criterion and review transfer, burden, safety and context change. Retirement triggers: criterion no longer met; unacceptable burden or safety event; invalid measure; loss of practical utility; plus 1 more.
Governance & provenance	Record HCE-TITAN-MOD-TEAM-001 · schema 0.2.0 · record version 1.0.0 · authority Titan · review 2026-10-24 · history append only with tombstones.

Table 96 — Capability Standard
Field	Value
ID	`MOD-TEAM-ORCHESTRATION`
Status	continued
Evidence	Strong
Risk	Low

Scope. Independent estimates, expertise maps, balanced contribution, dissent, closed-loop communication and after-action review.

Governing capability claim. Diagnose, make, repair or operate a bounded physical or procedural artifact to a declared quality and safety standard.

Claim boundary. This record does not establish general mechanical aptitude; safe competence with every tool; professional certification; repair beyond legal or technical authority.

Evidence contract. Direct: Artifact quality, task completion, defect rate, time, inspection result and assistance required. Retention: Repeat the technique after a delay without answer-bearing demonstration. Transfer: Apply the underlying process to a novel but comparable artifact, material or fault. The claim remains task-, route-, support- and version-specific; structural conformance does not prove empirical validity.

Table 97 — Capability Standard
Measure	Operational definition
Direct	Estimate accuracy, expertise coverage, closed-loop handoff or decision-memory completeness.
Retention	Team repeats the operating sequence after a delay.
Transfer	The sequence improves performance on a different team task.
Burden	Meeting time, participation imbalance, surveillance risk and dissent cost.

Protocol

Define the decision and expertise map.
Collect independent estimates before discussion.
Aggregate and expose assumptions.
Invite strongest dissent.
Close the loop on owner, action and review date.
Run an after-action review.

Level anchors

Table 98 — Capability Standard
Level	Module-specific anchor
L1	The team lacks expertise visibility, independent input or reliable decision memory.
L2	Uses the sequence with facilitation but participation or handoffs remain inconsistent.
L3	Meets the predeclared coverage, handoff or decision-quality criterion on the defined task.
L4	Retains the sequence and transfers it to a different team task.
L5	The team performs robustly across defined tasks and can onboard others without suppressing dissent.

Guardrail. Measure behaviour and outcomes, never identity proxies. Team metrics must not become punitive surveillance.

Worked exemplars

Independent estimates before discussion

ID. EX-INDEPENDENT-ESTIMATES

Narrow claim. Preserving independent judgments before group discussion can expose useful disagreement and reduce premature convergence.

Table 99 — Capability Standard
Test	Definition
Direct	Compare the quality and error of independent-first aggregation with the team’s declared baseline, while recording dissent coverage and coordination cost.
Retention	Repeat the process across at least two later team cycles.
Transfer	Use the method on a different decision or estimation task with the same ownership rules.
Guardrail	Protect psychological safety, attribution and the right to dissent. Do not infer an individual’s intelligence from the group result.

Evidence links

EVD-COLLECTIVE-INTELLIGENCE - Quantifying collective intelligence in human groups (Strong)
EVD-TEAMWORK-TRAINING - The Effectiveness of Teamwork Training on Teamwork Behaviors and Team Performance: A Systematic Review and Meta-Analysis of Controlled Interventions (Strong)

Governing capability claim. Design and operate repeatable workflows with verification, fallbacks, ownership and safe escalation.

Claim boundary. This record does not establish documentation as readiness; one successful drill as resilience; automation without accountable ownership; zero-failure claims.

Evidence contract. Direct: Process completion, error, recovery time, handoff quality or restored-artifact verification against a declared service target. Retention: Repeat the operating or recovery drill after a delay. Transfer: Apply the design method to a different bounded workflow or failure mode. The claim remains task-, route-, support- and version-specific; structural conformance does not prove empirical validity.

Making, Operations & Physical Systems

Purpose. Diagnose, make, maintain and operate physical or procedural systems with verification, fallbacks and safe escalation.

Family boundary. No competence claim for regulated or high-energy work outside training.

Making, Repair & Tool Intelligence

HCE Genome panel · MOD-TOOL-REPAIR · canonical registry record 1.0.0

HCE Genome panel · MOD-TOOL-REPAIR · canonical registry record 1.0.0
HCE GENOME FIELD	CANONICAL MODULE CONFIGURATION
Identity & status	Making, Repair & Tool Intelligence · family Making Tool Use Operations · actor scope Individual · status Operational.
Purpose	Diagnose, make, repair or operate a bounded physical or procedural artifact to a declared quality and safety standard.
Outcome & claim mode	Artifact quality, task completion, defect rate, time, inspection result and assistance required. Claim mode: Confirmatory; formation route: Acquisition.
Claim boundary	Population: Adults or configured teams for whom the declared task and route are appropriate. Conditions: predeclared task; versioned protocol and criterion; declared support and constraints. Excludes: general mechanical aptitude; safe competence with every tool; professional certification; repair beyond legal or technical authority.
Formation design	Primary mechanism: Procedural Compilation. Secondary: None declared. Patterns: worked example; deliberate practice; inspection and feedback; safe escalation.
Evidence & verification	Evidence grade: Emerging; lifecycle: Not Assessed; mode: Self Recorded. Baseline: Record performance before the formation protocol using the same or a matched scoring rule. Direct: Artifact quality, task completion, defect rate, time, inspection result and assistance required. Retention: Repeat the technique after a delay without answer-bearing demonstration. Transfer: Apply the underlying process to a novel but comparable artifact, material or fault.
Failure & interference	Failure modes: cosmetic success hiding defects; unsafe improvisation; inspection omission; tool misuse; scope creep. Interference plan: Measure and review: cosmetic success hiding defects, unsafe improvisation, inspection omission, tool misuse, scope creep
Safety, consent & agency	Risk: Variable. Voluntary participation with accessible withdrawal and no penalty Professional boundary: Use required PPE, isolation, food-safety, electrical, structural and professional rules. Stop where hidden hazards or regulated work exceed competence. Agency boundary: Capability support must not erase consent, accountable ownership or an appropriate fallback route.
Dependencies & ecology	Prerequisites: safe tool access; task specific instruction. Tools: None declared. Relationships: synergistic → MOD-OPERATIONAL-SYSTEMS; synergistic → MOD-LEARNING-TRANSFER; enables → MOD-ASSISTIVE-FIT.
Economics & burden	Equipment: Task- and context-dependent; prefer the lowest-burden safe route. Opportunity cost: Formation, assessment, maintenance and context-switching burden. Maintenance cost: M2–M3 by task frequency; requalify after long gaps for hazardous work.
Maintenance & retirement	Class M2; review due 90 days. Criterion: Repeat the declared direct criterion and review transfer, burden, safety and context change. Retirement triggers: criterion no longer met; unacceptable burden or safety event; invalid measure; loss of practical utility; plus 1 more.
Governance & provenance	Record HCE-TITAN-MOD-TOOL-001 · schema 0.2.0 · record version 1.0.0 · authority Titan · review 2026-10-24 · history append only with tombstones.

Table 101 — Capability Standard
Field	Value
ID	`MOD-TOOL-REPAIR`
Status	continued
Evidence	Emerging
Risk	Variable

Scope. Low-risk diagnosis, tool selection, maintenance, verified repair and safe escalation.

Exclusions. Tool-use plasticity does not establish broad repair intelligence or trade competence.

Measures

Table 102 — Capability Standard
Measure	Operational definition
Direct	Correct diagnosis, safe isolation, verified restoration or appropriate escalation.
Retention	Repeat the diagnostic and safety sequence after a delay.
Transfer	Diagnose a different low-risk failure using the same principles.
Burden	Near miss, injury, damage, cost and rework.

Protocol

Describe failure and stop condition.
Isolate energy and hazards.
Inspect and diagnose before replacing parts.
Choose the least destructive test.
Repair only within competence.
Verify, document and schedule maintenance.

Level anchors

Table 103 — Capability Standard
Level	Module-specific anchor
L1	Diagnosis, isolation or verification is incomplete on the selected low-risk task.
L2	Completes the task safely with guidance, checklist or supervision.
L3	Independently meets the diagnosis, isolation and verification criterion within competence.
L4	Retains the safety sequence and transfers it to a different low-risk failure.
L5	Performs robustly across defined low-risk systems and can teach the diagnostic sequence while escalating correctly.

Guardrail. Electricity, gas, structural systems, vehicles, pressure systems and other high-energy work belong with qualified professionals beyond your training.

Worked exemplars

Dim-sum fold or culinary technique progression

ID. EX-DIM-SUM-PROGRESSION

Narrow claim. A visible making skill can be improved through repeatable quality criteria, defect review and safe transfer to a related form.

Table 104 — Capability Standard
Test	Definition
Direct	For a dumpling fold, score seal integrity, weight consistency, shape consistency, defect rate and safe completion time across a fixed batch.
Retention	Repeat the same batch and scoring criteria after at least seven days.
Transfer	Use a predeclared different dough, filling consistency or fold while retaining food-safety and quality criteria.
Guardrail	Use food-safe ingredients and accessible tools. Knife, heat, allergen and commercial food work require appropriate competence, supervision and local rules.

Evidence links

Governing capability claim. Allocate work between human and AI components so the governed configuration improves a declared outcome over the better safe baseline without unacceptable defect, dependency, privacy or skill-loss cost.

Claim boundary. This record does not establish automatic superiority of human plus AI; delegation of accountability to a model; trust based on fluency; use of confidential data without authority; replacement of qualified professional review; retention of every manual skill regardless of purpose and cost.

Evidence contract. Direct: Compare human-only, AI-only where safe, and combined performance on the same predeclared quality, time, defect, provenance and verification-burden rubric. Retention: Repeat the human verification or fallback task after a delay without AI assistance to detect dependency or skill atrophy. Transfer: Apply the allocation, verification and escalation method to a materially different task without reusing confidential or answer-bearing content. The claim remains task-, route-, support- and version-specific; structural conformance does not prove empirical validity.

Operational Systems & Personal Infrastructure

HCE Genome panel · MOD-OPERATIONAL-SYSTEMS · canonical registry record 1.0.0

HCE Genome panel · MOD-OPERATIONAL-SYSTEMS · canonical registry record 1.0.0
HCE GENOME FIELD	CANONICAL MODULE CONFIGURATION
Identity & status	Operational Systems Design · family Making Tool Use Operations · actor scope Individual · status Operational.
Purpose	Design and operate repeatable workflows with verification, fallbacks, ownership and safe escalation.
Outcome & claim mode	Process completion, error, recovery time, handoff quality or restored-artifact verification against a declared service target. Claim mode: Confirmatory; formation route: Acquisition.
Claim boundary	Population: Adults or configured teams for whom the declared task and route are appropriate. Conditions: predeclared task; versioned protocol and criterion; declared support and constraints. Excludes: documentation as readiness; one successful drill as resilience; automation without accountable ownership; zero-failure claims.
Formation design	Primary mechanism: Support Environmental Redesign. Secondary: None declared. Patterns: checklist; backup and restore; fault containment; handoff.
Evidence & verification	Evidence grade: Moderate; lifecycle: Not Assessed; mode: Self Recorded. Baseline: Record performance before the formation protocol using the same or a matched scoring rule. Direct: Process completion, error, recovery time, handoff quality or restored-artifact verification against a declared service target. Retention: Repeat the operating or recovery drill after a delay. Transfer: Apply the design method to a different bounded workflow or failure mode.
Failure & interference	Failure modes: checklist theatre; stale backup; single point of failure; unclear ownership; unverified automation. Interference plan: Measure and review: checklist theatre, stale backup, single point of failure, unclear ownership, unverified automation
Safety, consent & agency	Risk: Low. Voluntary participation with accessible withdrawal and no penalty Professional boundary: Do not test destructive failure modes without backups, authority and containment. Protect credentials and confidential records. Agency boundary: Capability support must not erase consent, accountable ownership or an appropriate fallback route.
Dependencies & ecology	Prerequisites: defined service or artifact; authorised ownership. Tools: None declared. Relationships: synergistic → MOD-PROSPECTIVE-MEMORY; synergistic → MOD-TEAM-ORCHESTRATION; enables → MOD-HUMAN-AI-COMPLEMENTARITY.
Economics & burden	Equipment: Task- and context-dependent; prefer the lowest-burden safe route. Opportunity cost: Formation, assessment, maintenance and context-switching burden. Maintenance cost: M1–M2; schedule recurring restore, handoff and exception drills.
Maintenance & retirement	Class M1; review due 30 days. Criterion: Repeat the declared direct criterion and review transfer, burden, safety and context change. Retirement triggers: criterion no longer met; unacceptable burden or safety event; invalid measure; loss of practical utility; plus 1 more.
Governance & provenance	Record HCE-TITAN-MOD-OPS-001 · schema 0.2.0 · record version 1.0.0 · authority Titan · review 2026-10-24 · history append only with tombstones.

Table 106 — Capability Standard
Field	Value
ID	`MOD-OPERATIONAL-SYSTEMS`
Status	continued
Evidence	Moderate
Risk	Low

Scope. Triggers, owners, handoffs, checklists, fallbacks, tabletop tests and recovery learning.

Exclusions. A checklist cannot manufacture competence or transfer unchanged into every domain.

Measures

Table 107 — Capability Standard
Measure	Operational definition
Direct	Omission rate, handoff completeness, recovery time or failure rate.
Retention	The process still runs correctly after a defined interval.
Transfer	Adapt the design pattern to a different low-risk workflow.
Burden	Maintenance cost, alert fatigue, brittleness and workarounds.

Protocol

Define trigger and desired outcome.
Map the smallest reliable sequence.
Assign owner and handoffs.
Add a fallback for the likely failure.
Run a tabletop test.
Review failures and revise.

Level anchors

Table 108 — Capability Standard
Level	Module-specific anchor
L1	The process depends on memory or heroics and fails unpredictably.
L2	A documented sequence works with active prompting or close supervision.
L3	Meets the defined omission, handoff or recovery criterion in routine use.
L4	Retains reliability and transfers the design pattern to a different workflow.
L5	Runs robustly across defined disruptions and can be handed over without hidden knowledge.

Guardrail. High-risk systems still require qualified people, current procedures and applicable professional standards.

Worked exemplars

Backup-and-restore fire drill

ID. EX-BACKUP-RESTORE-FIRE-DRILL

Narrow claim. A backup is operationally useful only when a defined artifact can be restored and verified within the declared recovery target; one successful drill does not prove disaster readiness.

Table 109 — Capability Standard
Test	Definition
Direct	In a sandbox, choose a non-sensitive test artifact, declare a recovery-time target and verification checks, create the backup, remove only the sandbox working copy, restore it, then record success, recovery time, missing items, checksum or record-count match and whether the artifact opens or runs as expected.
Retention	Repeat the drill after at least thirty days using the same written procedure and a fresh test artifact, without relying on the original operator’s memory.
Transfer	Adapt the pattern to a different low-risk data type or workflow and predeclare its own restore checks and recovery target.
Guardrail	Never delete the sole copy or test destructively on production. Use synthetic or approved data, protect keys and personal information, preserve an independent known-good copy and obtain system-owner approval before touching shared infrastructure.

Evidence qualification. The current evidence supports checklist implementation and structured debriefing, not a universal backup technology or recovery-time promise.

Governing capability claim. Select, trial, adapt and review tools or supports against the intended task, context, burden and lived experience.

Claim boundary. This record does not establish availability as success; technical function as person–task fit; device use as personal worth; forced adoption.

Evidence contract. Direct: Task completion, accessibility, error, symptom-safe completion and assistance required with the proposed support. Retention: Sustained use and benefit at the scheduled review. Transfer: Benefit in another predeclared relevant context while preserving safety and burden criteria. The claim remains task-, route-, support- and version-specific; structural conformance does not prove empirical validity.

EVD-DEBRIEFING - Do team and individual debriefs enhance performance? A meta-analysis (Moderate)

Human–Technology Augmentation

Purpose. Select and govern assistive or AI tools so the combined system improves outcomes without losing safety, ownership or skill.

Family boundary. No assumption that adding AI or a device automatically improves performance.

Human–AI Complementarity & Reliance Calibration

HCE Genome panel · MOD-HUMAN-AI-COMPLEMENTARITY · canonical registry record 1.0.0

HCE Genome panel · MOD-HUMAN-AI-COMPLEMENTARITY · canonical registry record 1.0.0
HCE GENOME FIELD	CANONICAL MODULE CONFIGURATION
Identity & status	Human–AI Complementarity & Reliance Calibration · family Human–Technology Augmentation · actor scope Human–Technology Configuration · status Operational.
Purpose	Allocate work between human and AI components so the governed configuration improves a declared outcome over the better safe baseline without unacceptable defect, dependency, privacy or skill-loss cost.
Outcome & claim mode	Compare human-only, AI-only where safe, and combined performance on the same predeclared quality, time, defect, provenance and verification-burden rubric. Claim mode: Confirmatory; formation route: Augmentation.
Claim boundary	Population: Adults or configured teams for whom the declared task and route are appropriate. Conditions: predeclared task; versioned protocol and criterion; declared support and constraints. Excludes: automatic superiority of human plus AI; delegation of accountability to a model; trust based on fluency; use of confidential data without authority; plus 2 more.
Formation design	Primary mechanism: Reciprocal Adaptation. Secondary: Support Environmental Redesign. Patterns: reciprocal adaptation; baseline comparison; allocation and verification; fallback preservation; dissent preservation.
Evidence & verification	Evidence grade: Moderate; lifecycle: Not Assessed; mode: Instrumented. Baseline: Record performance before the formation protocol using the same or a matched scoring rule. Direct: Compare human-only, AI-only where safe, and combined performance on the same predeclared quality, time, defect, provenance and verification-burden rubric. Retention: Repeat the human verification or fallback task after a delay without AI assistance to detect dependency or skill atrophy. Transfer: Apply the allocation, verification and escalation method to a materially different task without reusing confidential or answer-bearing content.
Failure & interference	Failure modes: automation bias; over-trust from fluency; under-trust despite evidence; accountability diffusion; provenance loss; confidentiality breach; silent skill atrophy; vendor lock-in; conformity pressure; uncaught distribution shift. Interference plan: Measure and review: automation bias, over-trust from fluency, under-trust despite evidence, accountability diffusion, provenance loss
Safety, consent & agency	Risk: Variable. Voluntary participation with accessible withdrawal and no penalty Professional boundary: Protect confidential and regulated data, preserve accountable human ownership and escalate high-stakes disagreement to qualified review. Do not use a model’s confidence or eloquence as evidence. A combined system that improves speed while increasing unacceptable error, coercion, privacy loss or unreviewable dependency has failed. Agency boundary: Capability support must not erase consent, accountable ownership or an appropriate fallback route.
Dependencies & ecology	Prerequisites: domain literacy; source verification; task decomposition. Tools: None declared. Relationships: synergistic → MOD-ARGUMENTATION; synergistic → MOD-TEAM-ORCHESTRATION; enables → MOD-OPERATIONAL-SYSTEMS.
Economics & burden	Equipment: Tool fees may be modest or substantial; verification, privacy, switching and lock-in costs must be counted. Opportunity cost: The value proposition is comparative and non-compensatory: speed cannot erase safety or defect failure. Maintenance cost: M2 — review whenever models, prompts, data access, policies or task consequences change; revalidate after material model or workflow updates.
Maintenance & retirement	Class M2; review due 90 days. Criterion: Repeat the declared direct criterion and review transfer, burden, safety and context change. Retirement triggers: criterion no longer met; unacceptable burden or safety event; invalid measure; loss of practical utility; plus 1 more.
Governance & provenance	Record HCE-TITAN-MOD-HUMAN-AI-001 · schema 0.2.0 · record version 1.0.0 · authority Titan · review 2026-10-24 · history append only with tombstones.

Table 111 — Capability Standard
Field	Value
ID	`MOD-HUMAN-AI-COMPLEMENTARITY`
Status	promoted
Evidence	Moderate
Risk	Variable

Scope. Task allocation, verification, provenance, disagreement escalation, accountable override and skill-retention checks.

Exclusions. Human plus AI is not assumed to beat the better solo baseline.

Measures

Table 112 — Capability Standard
Measure	Operational definition
Direct	Combined quality, time, defect escape, disagreement catch and verification rate.
Retention	Repeat an unassisted trial after a defined assistance period.
Transfer	Use the allocation and verification method on a different task.
Burden	Privacy exposure, verification time, automation bias and skill loss.

Protocol

Define quality and safety criteria.
Record a human-only or best-known baseline.
Assign human, AI and joint work explicitly.
Verify sources and disagreements.
Compare combined performance with the best solo baseline.
Run periodic unassisted retention checks.

Level anchors

Table 113 — Capability Standard
Level	Module-specific anchor
L1	Uses assistance without a baseline, verification rule or ownership boundary.
L2	Applies a checklist with support but misses some defects or provenance gaps.
L3	The combined workflow meets the direct quality and verification criterion without exceeding the best-safe-baseline cost.
L4	Transfers the workflow to a new task and preserves predeclared unassisted performance.
L5	Reliably allocates and verifies across defined tasks, catches disagreement and can teach the governance method.

Guardrail. Protect confidential data, preserve accountable human ownership and escalate high-stakes disagreement to qualified review.

Worked exemplars

Human-only versus AI-assisted brief

ID. EX-HUMAN-AI-BRIEF

Narrow claim. A human-AI workflow is useful only when it beats the better safe baseline on a declared quality-and-cost rubric.

Table 114 — Capability Standard
Test	Definition
Direct	Compare human-only, AI-only when safe, and combined outputs for rubric quality, time, defect escape, disagreement catch and verification burden.
Retention	Repeat an unassisted human check after a delay to detect skill loss.
Transfer	Apply the allocation and verification method to a different task without reusing confidential material.
Guardrail	Protect private data, keep accountable human ownership and verify consequential claims against primary or authoritative sources.

Evidence links

EVD-HUMAN-AI-META — Vaccaro M, Almaatouq A, Malone TW. When combinations of humans and AI are useful: a systematic review and meta-analysis. Nat Hum Behav. 2024;8:2293–2303. DOI: 10.1038/s41562-024-02024-1. Across 106 experiments and 370 effect sizes, human–AI combinations were worse on average than the better solo component, with material variation by task type. The Titan implication is comparative testing, not a universal claim of harm or benefit. (Strong for the bounded meta-analytic claim)
EVD-GENAI-PRODUCTIVITY - Generative AI at Work (Moderate)
EVD-NIST-AI-RMF - Artificial Intelligence Risk Management Framework (AI RMF 1.0) (Framework)

EVD-TRUST-CALIBRATION-2024 - Carter OBJ, Loft S, Visser TAW. Meaningful Communication but not Superficial Anthropomorphism Facilitates Human-Automation Trust Calibration: The Human-Automation Trust Expectation Model (HATEM). Hum Factors. 2024;66(11):2485-2502. DOI: 10.1177/00187208231218156 (Moderate)

Reliance-calibration measure. Beside the module's existing measures, record appropriate reliance directly: the rate of accepting correct machine output, the rate of rejecting incorrect machine output, and the two failure directions separately — accepting when the machine is wrong, and rejecting when it is right. A single trust score conceals both.

Unaided-competence retest. Where a declared task is routinely performed with an AI system, the module requires a periodic support-off retest under decay band D4. This is the measure that detects silent skill atrophy, which the module already lists as a failure mode but did not previously test for.

Recovery after failure. Calibration is not established by performance while the system is reliable. Record how reliance behaves after the system has been observed to fail: durable disuse of a system that remains useful is a calibration failure in the opposite direction and is scored as one.

Distinctness finding. The frontier candidate 'human-AI trust and reliance calibration' (TFA-W1-011) was adjudicated against this module and found not to be distinct from it. It is recorded here as an evidence and measurement upgrade. No separate hybrid capability record was created, because doing so would have duplicated a named operational module.

Combination is not assumed superior, and the evidence says it often is not. A large meta-analysis of human-AI combinations found that on average they performed worse than the better of the human or the AI operating alone, with the direction varying by task type. This module therefore requires three measurements where a declared task can be performed by either party: human alone, system alone, and the pairing. A configuration that underperforms its own best component is a finding to record, not a failure to hide, and it is the expected result often enough that not measuring it would be negligent.

Assistive Technology & Person–Task Fit

HCE Genome panel · MOD-ASSISTIVE-FIT · canonical registry record 1.0.0

HCE Genome panel · MOD-ASSISTIVE-FIT · canonical registry record 1.0.0
HCE GENOME FIELD	CANONICAL MODULE CONFIGURATION
Identity & status	Assistive Technology & Person–Task Fit · family Human–Technology Augmentation · actor scope Individual · status Operational.
Purpose	Select, trial, adapt and review tools or supports against the intended task, context, burden and lived experience.
Outcome & claim mode	Task completion, accessibility, error, symptom-safe completion and assistance required with the proposed support. Claim mode: Confirmatory; formation route: Acquisition.
Claim boundary	Population: Adults or configured teams for whom the declared task and route are appropriate. Conditions: predeclared task; versioned protocol and criterion; declared support and constraints. Excludes: availability as success; technical function as person–task fit; device use as personal worth; forced adoption.
Formation design	Primary mechanism: Support Environmental Redesign. Secondary: None declared. Patterns: person task environment fit; trial and review; lowest burden route; abandonment as data.
Evidence & verification	Evidence grade: Moderate; lifecycle: Not Assessed; mode: Self Recorded. Baseline: Record performance before the formation protocol using the same or a matched scoring rule. Direct: Task completion, accessibility, error, symptom-safe completion and assistance required with the proposed support. Retention: Sustained use and benefit at the scheduled review. Transfer: Benefit in another predeclared relevant context while preserving safety and burden criteria.
Failure & interference	Failure modes: novelty bias; setup burden ignored; privacy cost ignored; lock in; abandonment treated as failure. Interference plan: Measure and review: novelty bias, setup burden ignored, privacy cost ignored, lock_in, abandonment treated as failure
Safety, consent & agency	Risk: Variable. Voluntary participation with accessible withdrawal and no penalty Professional boundary: Respect medical restrictions and lived experience. Pain, fatigue, unsafe performance, privacy harm or abandonment triggers review; discontinuing poor fit is valid data. Agency boundary: Capability support must not erase consent, accountable ownership or an appropriate fallback route.
Dependencies & ecology	Prerequisites: defined intended function; user consent and feedback. Tools: None declared. Relationships: synergistic → MOD-SENSORIMOTOR-RESILIENCE; synergistic → MOD-PROSPECTIVE-MEMORY; synergistic → MOD-HUMAN-AI-COMPLEMENTARITY.
Economics & burden	Equipment: Task- and context-dependent; prefer the lowest-burden safe route. Opportunity cost: Formation, assessment, maintenance and context-switching burden. Maintenance cost: M1–M2 during trial and after context, body, software or support changes.
Maintenance & retirement	Class M1; review due 30 days. Criterion: Repeat the declared direct criterion and review transfer, burden, safety and context change. Retirement triggers: criterion no longer met; unacceptable burden or safety event; invalid measure; loss of practical utility; plus 1 more.
Governance & provenance	Record HCE-TITAN-MOD-ASSISTIVE-001 · schema 0.2.0 · record version 1.0.0 · authority Titan · review 2026-10-24 · history append only with tombstones.

Table 116 — Capability Standard
Field	Value
ID	`MOD-ASSISTIVE-FIT`
Status	promoted
Evidence	Moderate
Risk	Variable

Scope. Select, trial, adapt and review tools against the intended task, context, burden and lived experience.

Exclusions. A device or substitution is not successful merely because it is available or technically functional.

Measures

Table 117 — Capability Standard
Measure	Operational definition
Direct	Task completion, accessibility, error, symptom-safe completion and assistance required.
Retention	Sustained use and benefit at the scheduled review.
Transfer	Benefit in another predeclared relevant context.
Burden	Setup, maintenance, fatigue, pain, stigma, privacy, cost and abandonment risk.

Protocol

Name the intended function.
Map person, task and environment constraints.
Trial the lowest-burden safe option.
Measure benefit and burden.
Review after real use.
Keep, modify, replace or abandon without penalty.

Level anchors

Table 118 — Capability Standard
Level	Module-specific anchor
L1	Current tool or pathway does not reliably support the intended task.
L2	Completes the task with substantial setup, support or burden.
L3	Meets the task criterion with acceptable burden in the primary context.
L4	Sustains benefit and transfers it to another relevant context.
L5	Maintains robust person–task fit across defined contexts and can document or teach the setup safely.

Guardrail. Respect medical restrictions and lived experience. Device trials must include burden and adverse-event review; abandonment is data, not failure.

Worked exemplars

Two-week assistive-technology fit trial

ID. EX-TWO-WEEK-ASSISTIVE-TRIAL

Narrow claim. A time-bounded real-use trial can compare task benefit with setup, symptom, privacy and abandonment burden; availability or novelty alone does not establish person-task fit.

Table 119 — Capability Standard
Test	Definition
Direct	Define one intended function and primary context, record at least three baseline attempts, then trial one safe option for fourteen days. Record task completion, time or error, assistance required, symptom-safe completion, setup time, maintenance, privacy concern and non-use days.
Retention	At days seven and fourteen, and again at a predeclared later review if the tool is kept, repeat the task measure and record sustained use, adaptation and burden rather than assuming continued benefit.
Transfer	Test the same option in one predeclared relevant secondary context while preserving the primary safety and burden criteria.
Guardrail	Respect medical restrictions and lived experience, use qualified fitting or training where required, protect device data and avoid unaffordable lock-in. Pain, fatigue, unsafe performance, privacy harm or abandonment triggers review; discontinuing a poor-fit tool is a valid result.

Evidence links

EVD-WHO-ASSISTIVE - Global report on assistive technology (Framework)
EVD-HUMAN-CENTRED-DESIGN - ISO 9241-210:2019 — Ergonomics of human-system interaction — Part 210: Human-centred design for interactive systems (Framework)

Revision 3 Claim-Equivalence Register

This register records the editorial migration from Revision 2 shorthand into the canonical HCE claim object. “Equivalent or narrower” means the new wording preserves the operational task while making population, conditions, support, duration, retention, transfer, safety and exclusions more explicit. No level threshold, direct-performance gate, retention rule, transfer rule, safety gate or module identity is changed by this revision. Full module-by-module source wording, governing wording, direct evidence, retention and transfer findings are supplied in the separate Revision 3 Claim-Equivalence Matrix.

Table 120 — Capability Standard
Module	Editorial migration	Claim direction	Operational criterion change
Sensory Discrimination & Substitution MOD-SENSORY-DISCRIMINATION	Legacy scope/exclusion shorthand reconciled to canonical Genome claim, boundary and evidence contract.	Equivalent or narrower	None
Spatial Simulation & Cognitive Maps MOD-SPATIAL-SIMULATION	Legacy scope/exclusion shorthand reconciled to canonical Genome claim, boundary and evidence contract.	Equivalent or narrower	None
Sensorimotor Intelligence & Fall Resilience MOD-SENSORIMOTOR-RESILIENCE	Legacy scope/exclusion shorthand reconciled to canonical Genome claim, boundary and evidence contract.	Equivalent or narrower	None
Learning, Memory & Skill Transfer MOD-LEARNING-TRANSFER	Legacy scope/exclusion shorthand reconciled to canonical Genome claim, boundary and evidence contract.	Equivalent or narrower	None
Metacognitive Calibration & Error Detection MOD-METACOGNITIVE-CALIBRATION	Legacy scope/exclusion shorthand reconciled to canonical Genome claim, boundary and evidence contract.	Equivalent or narrower	None
Creative Cognition & Insight MOD-CREATIVE-COGNITION	Legacy scope/exclusion shorthand reconciled to canonical Genome claim, boundary and evidence contract.	Equivalent or narrower	None
Prospective Memory & External Cues MOD-PROSPECTIVE-MEMORY	Legacy scope/exclusion shorthand reconciled to canonical Genome claim, boundary and evidence contract.	Equivalent or narrower	None
Quantitative Intelligence & Forecasting MOD-FORECASTING	Legacy scope/exclusion shorthand reconciled to canonical Genome claim, boundary and evidence contract.	Equivalent or narrower	None
Decision, Causal & Experimental Intelligence MOD-CAUSAL-EXPERIMENTAL	Legacy scope/exclusion shorthand reconciled to canonical Genome claim, boundary and evidence contract.	Equivalent or narrower	None
Argumentation & Dialectical Testing MOD-ARGUMENTATION	Legacy scope/exclusion shorthand reconciled to canonical Genome claim, boundary and evidence contract.	Equivalent or narrower	None
Listening, Teaching & Negotiation MOD-TEACHING-NEGOTIATION	Legacy scope/exclusion shorthand reconciled to canonical Genome claim, boundary and evidence contract.	Equivalent or narrower	None
Collective Intelligence & Team Orchestration MOD-TEAM-ORCHESTRATION	Legacy scope/exclusion shorthand reconciled to canonical Genome claim, boundary and evidence contract.	Equivalent or narrower	None
Making, Repair & Tool Intelligence MOD-TOOL-REPAIR	Legacy scope/exclusion shorthand reconciled to canonical Genome claim, boundary and evidence contract.	Equivalent or narrower	None
Operational Systems Design MOD-OPERATIONAL-SYSTEMS	Legacy scope/exclusion shorthand reconciled to canonical Genome claim, boundary and evidence contract.	Equivalent or narrower	None
Human–AI Complementarity & Reliance Calibration MOD-HUMAN-AI-COMPLEMENTARITY	Legacy scope/exclusion shorthand reconciled to canonical Genome claim, boundary and evidence contract.	Equivalent or narrower	None
Assistive Technology & Person–Task Fit MOD-ASSISTIVE-FIT	Legacy scope/exclusion shorthand reconciled to canonical Genome claim, boundary and evidence contract.	Equivalent or narrower	None

Release decision. All sixteen modules are accepted as editorially reconciled at Genome record version 1.0.0. Any future substantive change to a task, criterion, safety boundary, support configuration, retention window or transfer class requires a version increment, migration note and regression test.

Wave 1 verified evidence ledger

Revision 5 verified source identity, publication status, study population, sample size, design and direct functional outcome for all eight Wave 1 evidence records. Retention, transfer and adverse-event reporting were recorded as absent or limited where the primary source did not establish them. “Verified” means that the manuscript accurately represents the cited study; it does not make the study definitive or convert a Candidate or Observatory item into an operational module.

EVD-CALENDAR-CALCULATION — Calendar Engine

Verified grade: Moderate. Ward et al. (2019) trained sequence-space synaesthetes and controls for approximately one hour over two weeks. At final testing, participants selected weekdays for dates drawn from several thousand possibilities in roughly ten seconds with about 80% accuracy. Direct acquisition is supported; long-term retention and far transfer were not established. DOI: 10.1016/j.cognition.2018.11.012

EVD-VISUAL-HYPERACUITY — Visual Hyperacuity

Verified grade: Moderate. Snell et al. (2015) trained seven adults on Vernier acuity for 7,500 trials. Performance improved on the trained task and orientation, with little evidence of transfer to the untrained orientation. Direct task learning is supported; general visual transfer is not. DOI: 10.1371/journal.pone.0145770

EVD-OLFACTORY-SPATIAL-NAV — Olfactory Landmark Navigation

Verified grade: Emerging–Moderate. Hamburger and Knauff (2019) tested 24 participants in the main experiment and found that odours could function as wayfinding landmarks. Schwarz and Hamburger (2024) tested 52 participants and reported above-chance olfactory route decisions immediately and one month later. Ordinary real-world navigation remains unsupported. DOIs: 10.1111/cogs.12798; 10.1007/s10339-023-01169-7

EVD-AUDITORY-SPATIAL-REMAP — Auditory Spatial Remapping

Verified grade: Moderate. Hofman et al. (1998) altered adult listeners’ pinna cues. Elevation localisation initially deteriorated and was progressively reacquired under the altered mapping. Permanent remapping, broad transfer and movement safety were not established. DOI: 10.1038/nn0998_417

EVD-NOVEL-CUE-INTEGRATION — Capability Maturity Vector

Verified grade: Moderate–Limited. Karni and Zaidel (2026) tested 20 adults. Combining a synthetic auditory self-motion cue with vestibular stimulation improved heading discrimination, but integration was suboptimal and the synthetic cue was underweighted. DOI: 10.1016/j.isci.2026.114885

EVD-SYNTHETIC-SELF-MOTION-2026 — Candidate Lab

Verified grade: Emerging. The same study supports immediate combined-cue benefit in one discrimination task. It did not test training, automaticity, device-off retention, a clinical population or navigation. Candidate status is retained. DOI: 10.1016/j.isci.2026.114885

EVD-PUPIL-BIOFEEDBACK — Research Observatory

Verified grade: Moderate. De Gee et al. (2024) randomly assigned 28 participants to veridical pupil biofeedback and 28 to control feedback in Experiment 1A. Three-day training improved pupil up- and down-regulation, including no-feedback trials after training. Durable transfer and clinical value remain unresolved. DOI: 10.1038/s41562-023-01729-z

EVD-N1-INSIGHT — Research Observatory

Verified grade: Emerging–Moderate. Lacaux et al. (2021) studied 103 participants. Hidden-rule discovery occurred in 20 of 24 analysed N1 participants, 15 of 49 wake participants and 2 of 14 N2 participants. This was one bounded laboratory task; replication, transfer and practical induction remain unresolved. DOI: 10.1126/sciadv.abj5866

Revision 6 primary-source register

Ward, J., Thompson-Lake, D., Ely, R. & Kaminski, F. (2019). Synaesthetes show advantages in savant skill acquisition: Training calendar calculation in sequence-space synaesthesia. Cognition, 187, 100–110. https://doi.org/10.1016/j.cognition.2018.11.012

Snell, N., Kattner, F., Rokers, B. & Green, C. S. (2015). Orientation transfer in Vernier and stereoacuity training. PLOS ONE, 10(12), e0145770. https://doi.org/10.1371/journal.pone.0145770

Hamburger, K. & Knauff, M. (2019). Odors can serve as landmarks in human wayfinding. Cognitive Science, 43, e12798. https://doi.org/10.1111/cogs.12798

Schwarz, M. & Hamburger, K. (2024). Memory effects of visual and olfactory landmark information in human wayfinding. Cognitive Processing, 25, 37–51. https://doi.org/10.1007/s10339-023-01169-7

Hofman, P. M., Van Riswick, J. G. A. & Van Opstal, A. J. (1998). Relearning sound localization with new ears. Nature Neuroscience, 1, 417–421. https://doi.org/10.1038/nn0998_417

Karni, R. & Zaidel, A. (2026). Augmentation of self-motion perception with synthetic auditory cues. iScience, 29(3), 114885. https://doi.org/10.1016/j.isci.2026.114885

de Gee, J. W. et al. (2024). Self-regulating arousal via pupil-based biofeedback. Nature Human Behaviour, 8, 43–62. https://doi.org/10.1038/s41562-023-01729-z

Lacaux, C. et al. (2021). Sleep onset is a creative sweet spot. Science Advances, 7(50), eabj5866. https://doi.org/10.1126/sciadv.abj5866

Research clarifications without promotion

Empathy-training studies exist, including role-specific intervention trials and meta-analyses (PubMed record). Titan does not infer a broad, durable empathy trait from that literature. A future route must target bounded empathic communication, blinded behavioural ratings, recipient outcomes, retention and transfer.
Deception-detection training studies also exist, but aggregate effects are modest and asymmetric (meta-analysis). Titan therefore retains reliable general lie detection as excluded and redirects source/claim verification to Argumentation and the unscored Information Verification candidate.

Research existence is not module-grade evidence. The exclusion rests on claim breadth, ecological validity, retention and transfer - not on pretending that no training study has been published.

Explicit exclusions and honest redirects

Generic brain games raise IQ or provide broad far transfer. Nearest honest home: task-specific retention and predeclared transfer in Learning, Memory & Skill Transfer (MOD-LEARNING-TRANSFER).
Extreme speed reading preserves full comprehension. Nearest honest home: comprehension-controlled reading practice in Learning, Memory & Skill Transfer (MOD-LEARNING-TRANSFER).
Microexpressions provide reliable lie detection. Nearest honest home: source and claim verification in the Information Verification candidate, with Argumentation & Dialectical Testing (MOD-ARGUMENTATION) as the current operational route.
Photographic memory can be broadly trained. Nearest honest home: mnemonics, retrieval, retention and honest transfer testing in Learning, Memory & Skill Transfer (MOD-LEARNING-TRANSFER).
Meditation is a risk-free universal cognitive enhancer. Nearest honest home: a declared task outcome under the Stress Recovery candidate, with clinical symptoms routed to qualified care.
Titan excludes any claim that DIY neurostimulation is a validated general enhancement route. The topic remains only in the unscored Consumer Neurostimulation Observatory record (OBS-NEUROSTIMULATION).
Cold, hypoxia, fasting, pain or breath-hold tolerance demonstrates superior character. Nearest honest home: clinician-approved functional work in Sensorimotor Intelligence & Fall Resilience (MOD-SENSORIMOTOR-RESILIENCE), never a toughness score.
Polyphasic sleep schedules, sleep compression or deliberate sleep restriction as a performance superpower. Nearest honest home: the unscored Sleep Regularity & Recovery candidate (MOD-SLEEP-REGULARITY), which preserves adequate sleep opportunity and routes clinical sleep concerns to professionals.

Versioning and expiry

Record the registry, module, evidence review and Cockpit state versions beside every result.
A changed protocol or threshold creates a new assessment context; it does not rewrite a historical result.
Display overdue or expired evidence as historical rather than current.
Retire measures that do not earn their burden while preserving provenance and missing-data labels.