Structural Mapping Transfer¶

Transfer an explanation, model, or intervention between domains by mapping relational structure, testing mismatches, adapting only warranted elements, and validating the result in the target.

Version: v1 · 2026-08-24 · History
Solution archetype #: 1024
Problem family: Uncertainty, Evidence & Inference Failure
Problem subfamily: Explanatory Hypothesis, Pattern & Case Reasoning

Essence¶

Structural Mapping Transfer turns analogy from an intuition into a governed transfer process. It identifies a source whose relational structure may illuminate a target, represents both systems independently, makes correspondences and absences explicit, and permits only the portion of a source explanation or intervention that survives invariant, difference, and target-evidence tests. The archetype creates leverage without granting source prestige a blank check.

Canonical formula: source_structure + target_problem + correspondence_map + invariant_and_difference_tests + bounded_transfer_rule + target_validation + revision_loop -> warranted_cross_domain_transfer

When This Archetype Applies¶

Partial catalog groundingSome structural conditions are represented by existing abstractions, but no sufficient condition set is fully represented.

Diagnostic problem

A source and target may share a transferable relational structure despite different surface forms, but a false analogy could import the wrong causal direction, roles, scale, or constraints.

What this problem means

Cross-domain reuse faces a double hazard. Without abstraction, teams miss transferable structures and repeatedly reinvent solutions. With uncontrolled abstraction, they equate objects that look alike while ignoring causal direction, authority, scale, semantics, and constraints. The result is either lost learning or negative transfer. A valid intervention must preserve source leverage, target independence, difference visibility, and the right to reject the analogy.

Applicability expression2 distinct conditions

Shared relational structureandTransfer-invalidating differences

Algebraic1∧2

groundedpartly groundedopen

2 conditions, all required.

2Required in every casenumbered 1–2

These hold no matter which pattern applies.

1

Shared relational structure · grounded

A source and target may share relational structure despite different surface objects or vocabulary.

primeAnalogy— Transfer structure between domains.

and — also required

2

Transfer-invalidating differences · open

Differences in causal direction, authority, scale, semantics, or constraints could invalidate the transfer.

Other requirements and context (1)

Why these sit outside the expression

Solution feasibility — it describes whether the intervention can work, not whether the diagnostic problem exists.

Solution feasibilityThe proposed role and relation mapping can be tested against both domains.

Coverage

1 of 2 conditions grounded · 1 open.

Read the methodology Download the trigger-logic data

When to Use This Archetype¶

Use this archetype when deep structural similarity may unlock an explanation, model, or intervention across domains and when a false analogy would be costly. It is especially valuable when surface language hides useful correspondence, when teams disagree about the validity of a transfer, or when target adaptation threatens to erase the relation that made the source effective. Do not use it for decorative metaphor, ordinary template reuse, or a relation inventory inside one system.

Structural Problem¶

Cross-domain reuse faces a double hazard. Without abstraction, teams miss transferable structures and repeatedly reinvent solutions. With uncontrolled abstraction, they equate objects that look alike while ignoring causal direction, authority, scale, semantics, and constraints. The result is either lost learning or negative transfer. A valid intervention must preserve source leverage, target independence, difference visibility, and the right to reject the analogy.

Intervention Logic¶

Begin by modeling source and target independently. Extract a relational signature from the source and create a typed correspondence map to the target, including absent and uncertain matches. Test required invariants and search for disanalogies. State a bounded transfer rule, adapt eligible elements to target constraints, and validate predictions and outcomes in a reversible target trial. Version the mapping and exit when evidence shows that the warrant does not hold.

Key Components¶

Component	Description
Source Problem and Solution Model ↗	Defines the source problem, intervention, outcomes, and causal or relational structure from which transfer is proposed. Operationally, this component must be represented in a form that decision makers can inspect and update. Record source evidence, assumptions, scope, failure modes, and the level at which the source structure operates. Its owner, evidence source, confidence, and update cadence should be explicit; otherwise the component becomes a decorative label rather than load-bearing structure. The component connects to the archetype by anchoring transfer in an inspected source rather than a memorable story. It is required because practitioners transfer an impression without knowing which source relationships produced success Reviewers should test it with ordinary, boundary, missing-data, and adversarial cases instead of assuming that a plausible description will behave correctly in use.
Target Problem Model ↗	Represents target actors, states, constraints, relations, outcomes, and unresolved design obligations independently of the source. Operationally, this component must be represented in a form that decision makers can inspect and update. Build the target model before forcing correspondences; include local vocabulary and stakeholder evidence. Its owner, evidence source, confidence, and update cadence should be explicit; otherwise the component becomes a decorative label rather than load-bearing structure. The component connects to the archetype by preventing the source from silently defining the target. It is required because target evidence is selected only to confirm the analogy Reviewers should test it with ordinary, boundary, missing-data, and adversarial cases instead of assuming that a plausible description will behave correctly in use.
Relational Structure Signature ↗	Extracts the source pattern at the level of roles, dependencies, causal direction, feedback, sequence, and constraint. Operationally, this component must be represented in a form that decision makers can inspect and update. Distinguish load-bearing relations from incidental objects, labels, technology, and cultural presentation. Its owner, evidence source, confidence, and update cadence should be explicit; otherwise the component becomes a decorative label rather than load-bearing structure. The component connects to the archetype by making the transferable object explicit. It is required because surface resemblance substitutes for structural equivalence Reviewers should test it with ordinary, boundary, missing-data, and adversarial cases instead of assuming that a plausible description will behave correctly in use.
Typed Correspondence Map ↗	Pairs source roles, relations, states, and transformations with proposed target counterparts and records correspondence type and confidence. Operationally, this component must be represented in a form that decision makers can inspect and update. Permit one-to-many, many-to-one, absent, and uncertain mappings rather than forcing a neat bijection. Its owner, evidence source, confidence, and update cadence should be explicit; otherwise the component becomes a decorative label rather than load-bearing structure. The component connects to the archetype by exposing what is claimed to carry across domains. It is required because critical correspondences remain implicit and cannot be challenged Reviewers should test it with ordinary, boundary, missing-data, and adversarial cases instead of assuming that a plausible description will behave correctly in use.
Invariant Preservation Test ↗	Checks whether the relations necessary for the source intervention remain true or can be recreated in the target. Operationally, this component must be represented in a form that decision makers can inspect and update. Test ordering, conservation, authority, feedback polarity, information access, reversibility, and resource constraints as applicable. Its owner, evidence source, confidence, and update cadence should be explicit; otherwise the component becomes a decorative label rather than load-bearing structure. The component connects to the archetype by separating valid transfer from loose analogy. It is required because a solution is imported after its enabling conditions have disappeared Reviewers should test it with ordinary, boundary, missing-data, and adversarial cases instead of assuming that a plausible description will behave correctly in use.
Disanalogy and Difference Register ↗	Records unmapped elements, reversed relations, scale changes, institutional differences, value conflicts, and new failure pathways. Operationally, this component must be represented in a form that decision makers can inspect and update. Assign each difference an effect on validity, adaptation need, test plan, and stop condition. Its owner, evidence source, confidence, and update cadence should be explicit; otherwise the component becomes a decorative label rather than load-bearing structure. The component connects to the archetype by preventing persuasive similarities from hiding decisive mismatches. It is required because negative transfer appears only after implementation harm Reviewers should test it with ordinary, boundary, missing-data, and adversarial cases instead of assuming that a plausible description will behave correctly in use.
Bounded Transfer Rule ↗	States which source claims or intervention elements may transfer, under what conditions, and which may not. Operationally, this component must be represented in a form that decision makers can inspect and update. Express the rule as a conditional warrant with scope, exclusions, confidence, and expiry or review triggers. Its owner, evidence source, confidence, and update cadence should be explicit; otherwise the component becomes a decorative label rather than load-bearing structure. The component connects to the archetype by converting comparison into an actionable but limited decision. It is required because the map remains descriptive or becomes an unrestricted license to copy Reviewers should test it with ordinary, boundary, missing-data, and adversarial cases instead of assuming that a plausible description will behave correctly in use.
Target Adaptation Design ↗	Modifies transferable elements to fit target constraints without destroying the mapped invariants. Operationally, this component must be represented in a form that decision makers can inspect and update. Track adaptations separately from source elements and explain why each is necessary. Its owner, evidence source, confidence, and update cadence should be explicit; otherwise the component becomes a decorative label rather than load-bearing structure. The component connects to the archetype by making transfer responsive to real target differences. It is required because literal copying replaces structurally faithful adaptation Reviewers should test it with ordinary, boundary, missing-data, and adversarial cases instead of assuming that a plausible description will behave correctly in use.
Target Validation Protocol ↗	Tests transferred predictions, behavior, safety, and outcomes in the target with comparison cases and explicit failure criteria. Operationally, this component must be represented in a form that decision makers can inspect and update. Begin with bounded, reversible trials when uncertainty or consequence is high. Its owner, evidence source, confidence, and update cadence should be explicit; otherwise the component becomes a decorative label rather than load-bearing structure. The component connects to the archetype by requiring target evidence rather than source prestige. It is required because apparent plausibility is treated as demonstrated effectiveness Reviewers should test it with ordinary, boundary, missing-data, and adversarial cases instead of assuming that a plausible description will behave correctly in use.
Mapping Revision and Exit Log ↗	Versions correspondences, records evidence, revises confidence, and abandons mappings that fail. Operationally, this component must be represented in a form that decision makers can inspect and update. Preserve rejected correspondences and reasons so later teams do not repeat the same false analogy. Its owner, evidence source, confidence, and update cadence should be explicit; otherwise the component becomes a decorative label rather than load-bearing structure. The component connects to the archetype by closing the learning loop and governing uncertainty. It is required because mappings harden into doctrine despite accumulating counterevidence Reviewers should test it with ordinary, boundary, missing-data, and adversarial cases instead of assuming that a plausible description will behave correctly in use.

Common Mechanisms¶

Mechanism	Description
Structure-First Abstraction ↗	Rewrites source and target cases in role-and-relation form before comparing domain objects or vocabulary. Use it when surface details are salient enough to bias similarity judgments Implementation should expose inputs, thresholds or transformation rules, expected outputs, responsible actors, and evidence of performance. Retain causal and normative context that materially conditions the relations. This mechanism is not the archetype by itself. It instantiates the broader intervention only when connected to the complete component set, monitoring, exception handling, and revision logic. This extraction method supplies the relational signature but is not the full transfer lifecycle.
Systematic Analogy Table ↗	Compares source and target elements, relations, causal directions, multiplicities, and confidence in a reviewable matrix. Use it when multiple experts need to inspect or contest a proposed analogy Implementation should expose inputs, thresholds or transformation rules, expected outputs, responsible actors, and evidence of performance. Include explicit no-match and uncertain cells rather than manufacturing completeness. This mechanism is not the archetype by itself. It instantiates the broader intervention only when connected to the complete component set, monitoring, exception handling, and revision logic. The table implements correspondence mapping; validation and adaptation remain separate obligations.
Graph and Causal-Model Alignment ↗	Aligns nodes, edges, directions, feedback loops, and intervention points across source and target representations. Use it when causal or network topology is the proposed transferable structure Implementation should expose inputs, thresholds or transformation rules, expected outputs, responsible actors, and evidence of performance. Similar topology does not establish identical semantics, parameters, or intervention effects. This mechanism is not the archetype by itself. It instantiates the broader intervention only when connected to the complete component set, monitoring, exception handling, and revision logic. Model alignment is one technical mechanism within the archetype.
Disanalogy Stress Test ↗	Searches deliberately for target differences that reverse, weaken, or invalidate the mapped intervention logic. Use it when the analogy is compelling, high stakes, or supported mainly by advocates Implementation should expose inputs, thresholds or transformation rules, expected outputs, responsible actors, and evidence of performance. Reward valid rejection and revision rather than treating objections as implementation resistance. This mechanism is not the archetype by itself. It instantiates the broader intervention only when connected to the complete component set, monitoring, exception handling, and revision logic. Counterexample search tests the transfer warrant but does not itself design the target adaptation.
Counterfactual Correspondence Probe ↗	Varies a proposed source-target correspondence and predicts how target outcomes should change if the mapping is real. Use it when observational resemblance cannot distinguish competing maps Implementation should expose inputs, thresholds or transformation rules, expected outputs, responsible actors, and evidence of performance. State auxiliary assumptions and avoid claiming causal proof from one probe. This mechanism is not the archetype by itself. It instantiates the broader intervention only when connected to the complete component set, monitoring, exception handling, and revision logic. The probe contributes discriminating evidence to the validation protocol.
Invariant Checklist and Gate ↗	Makes required source conditions explicit and blocks transfer when critical invariants are absent or unverified. Use it when an intervention depends on ordering, authority, conservation, feedback, or resource relations Implementation should expose inputs, thresholds or transformation rules, expected outputs, responsible actors, and evidence of performance. Do not turn the checklist into a universal template; derive invariants from the inspected source. This mechanism is not the archetype by itself. It instantiates the broader intervention only when connected to the complete component set, monitoring, exception handling, and revision logic. A gate is a control mechanism, not a substitute for modeling source and target.
Bounded Analogical Pilot ↗	Tests an adapted source intervention in a limited target setting with comparison, monitoring, and rollback. Use it when target validation is possible but uncertainty makes full transfer premature Implementation should expose inputs, thresholds or transformation rules, expected outputs, responsible actors, and evidence of performance. Define harm limits, learning goals, representativeness limits, and exit authority before launch. This mechanism is not the archetype by itself. It instantiates the broader intervention only when connected to the complete component set, monitoring, exception handling, and revision logic. Piloting validates one transfer rule; it is not general transfer scaffolding.
Mapping Version Control ↗	Records correspondence changes, evidence, confidence, decisions, and deprecated mappings over time. Use it when source knowledge or target conditions evolve during use Implementation should expose inputs, thresholds or transformation rules, expected outputs, responsible actors, and evidence of performance. Version both models and preserve the decision consequences of changed mappings. This mechanism is not the archetype by itself. It instantiates the broader intervention only when connected to the complete component set, monitoring, exception handling, and revision logic. Version control supports auditability but does not determine structural validity.

Parameter / Tuning Dimensions¶

abstraction_level¶

Controls how far source and target are reduced from domain particulars into relational structure. Low settings create missed deep correspondence because surface vocabulary dominates; high settings create decisive context, semantics, and values are erased. Tune against prediction accuracy, expert challenge, and whether preserved relations actually govern outcomes and record the rationale.

correspondence_strictness¶

Sets the evidence required to accept a proposed source-target match. Low settings create false mappings and negative transfer; high settings create useful partial analogies are rejected before bounded testing. Tune against consequence severity, reversibility, and discriminating target tests and record the rationale.

transfer_scope¶

Sets how much of the source explanation or intervention may be carried into the target. Low settings create fragmented benefit and excessive reinvention; high settings create unsupported wholesale copying. Tune against invariant coverage and disanalogy register severity and record the rationale.

validation_intensity¶

Controls target testing depth, comparison coverage, and monitoring duration. Low settings create plausible but false transfer survives; high settings create validation cost exceeds the value or timeliness of reuse. Tune against stakes, uncertainty, observability, and reversibility and record the rationale.

adaptation_freedom¶

Controls how much target modification is allowed before the source analogy no longer provides a warrant. Low settings create brittle literal copying; high settings create the intervention changes so much that source evidence no longer applies. Tune against invariant preservation and target outcome comparisons and record the rationale.

Invariants to Preserve¶

relation_before_resemblance¶

Transfer claims must be based on explicit role and relation correspondence rather than surface similarity. Preserve it by requiring a typed correspondence map and structural signature A violation is indicated by persuasive analogies lack inspectable relational claims

target_independence¶

The target problem must be modeled in its own evidence and vocabulary before alignment. Preserve it by separating target elicitation from correspondence construction A violation is indicated by all target facts are expressed only through source categories

difference_visibility¶

Material disanalogies and unmatched elements remain visible in decisions. Preserve it by maintaining an owned difference register with stop conditions A violation is indicated by exceptions are reframed as implementation noise

bounded_warrant¶

Source evidence warrants only the transferred elements whose conditions survive in the target. Preserve it by declaring scope, exclusions, confidence, and expiry A violation is indicated by source success is cited as proof of general target effectiveness

target_evidence_primacy¶

Continued use depends on target validation and revision. Preserve it by monitoring target predictions, harms, and outcomes with exit authority A violation is indicated by counterevidence is dismissed because the source model is prestigious

Target Outcomes¶

A successful application produces a reviewable correspondence map, explicit transfer boundaries, target-adapted intervention elements, evidence that transferred predictions or mechanisms operate in the target, and an auditable record of differences and revisions. It should reduce reinvention and accelerate insight while lowering the rate and consequence of false analogy. The outcome is warranted reuse, not proof that two domains are identical.

Tradeoffs¶

reuse_speed_vs_validation¶

Structural transfer can accelerate discovery, but correspondence testing consumes time and evidence. The design should not pretend this tension disappears. Use tiering validation by consequence and reversibility while preserving mandatory invariant gates and make the selected position visible to affected actors.

abstraction_vs_context¶

Abstraction reveals deep similarity while risking removal of target-specific meaning and constraints. The design should not pretend this tension disappears. Use maintaining linked source, target, and correspondence models rather than one universal diagram and make the selected position visible to affected actors.

simplicity_vs_difference¶

A compact analogy aids coordination but can hide one-to-many mappings and exceptions. The design should not pretend this tension disappears. Use using a simple public summary backed by a complete difference register and make the selected position visible to affected actors.

exploration_vs_false_transfer¶

Permissive analogy search yields creative options and more spurious matches. The design should not pretend this tension disappears. Use separating low-cost hypothesis generation from high-evidence adoption gates and make the selected position visible to affected actors.

source_leverage_vs_target_agency¶

Source expertise can accelerate design while crowding out target knowledge and affected stakeholders. The design should not pretend this tension disappears. Use giving target evidence and local challenge explicit decision rights and make the selected position visible to affected actors.

Failure Modes¶

Surface Similarity Substitution¶

Similar objects or vocabulary are mistaken for equivalent relations. Detect it through reviewers cannot state the mapped causal or functional relations Respond by rebuild the comparison at the role-and-relation level or abandon it

Forced Correspondence¶

Every source element is assigned a target even when no valid counterpart exists. Detect it through no-match and uncertainty categories are absent from the map Respond by permit partial mappings and downgrade or stop the transfer

Unmapped Constraint Failure¶

A target legal, physical, social, or resource constraint invalidates the source intervention. Detect it through pilot failures cluster around a target condition missing from the source model Respond by add the constraint, redesign the adaptation, and rerun invariant tests

Adaptation Erases the Invariant¶

Target customization changes the relation that made the source solution work. Detect it through the adapted intervention no longer satisfies a declared invariant Respond by restore the invariant, select another source, or treat the design as novel

Mapping Lock-In¶

A productive initial analogy becomes an unquestioned worldview. Detect it through counterevidence changes parameters but never the correspondence itself Respond by schedule competing-map review and authorize explicit deprecation

Neighbor Distinctions¶

Relation Mapping¶

Makes associations, dependencies, obligations, or influence paths visible inside a system.

Boundary rule: Use Relation Mapping when the design object is the relation structure of one target; use Structural Mapping Transfer when a source structure is being carried to a different target and transfer validity must be tested.