Skip to content

Alternative Hypothesis Generation

Before treating a conclusion as settled, generate credible alternative explanations and identify the evidence that would distinguish them.

Version
v1 · 2026-08-24 · History
Solution archetype #
45
Problem family
Uncertainty, Evidence & Inference Failure
Problem subfamily
Explanatory Hypothesis, Pattern & Case Reasoning

Gap-fill role

This draft directly addresses the zero-any accepted-prime target epistemic_humility from queue position 22 of scaled_gap_fill_batch_003_queue.yaml. It operationalizes humility as a rival-generation discipline: before confidence hardens, the reasoner must name plausible alternatives, identify discriminating evidence, and keep confidence proportional to unresolved rivals.

Pre-draft disposition conclusion

Disposition: drafted_full_archetype. The candidate has close neighbors in observational_equivalence_resolution, counterexample_search, bias_specific_decision_audit, bayesian_belief_updating, cautious_pattern_completion, assumption_stress_testing, causal_mechanism_mapping, and the pilot counter_narrative_probe mechanism. None of these clearly absorbs the pattern as a complete archetype, and no binding alias or duplicate-merge directive was found.

When This Archetype Applies

No catalog groundingNone of the structural conditions is currently represented by an accepted prime or domain-specific abstraction.

Diagnostic problem

A favored explanation is treated as settled before credible alternatives are generated, compared, or tested even though the available evidence underdetermines the conclusion.

What this problem means

A reasoner, team, institution, or model converges on a favored explanation before credible alternatives have been named, compared, or tested. The conclusion may be narratively coherent, emotionally satisfying, institutionally convenient, or statistically salient, yet still underdetermined by the available evidence. Without a rival-hypothesis step, uncertainty is hidden and confidence rises faster than warrant.

Applicability expression1 distinct condition

Premature explanation lock-in
Algebraic1

groundedpartly groundedopen

1 condition, all required.

1Required in every casenumbered 1–1

These hold no matter which pattern applies.

1

Premature explanation lock-in · open

One explanation becomes dominant before evidence has ruled out plausible rivals.

Other requirements and context (5)

Why these sit outside the expression

Supporting context — it may accompany or help interpret the situation, but it is not a load-bearing condition in a sufficient diagnostic set.

Application gate — it governs whether applying the archetype is appropriate or material, rather than defining the structural problem itself.

  • Supporting contextThe same observations could plausibly arise from multiple mechanisms, causes, actors, or contexts.

  • Application gateThe conclusion is high-stakes, irreversible, reputationally damaging, safety-relevant, or costly.

  • Supporting contextThe reasoning environment rewards decisiveness, coherence, authority, or speed more than uncertainty disclosure.

  • Supporting contextEvidence is sparse, noisy, incomplete, selected, or vulnerable to measurement artifact.

  • Supporting contextExperts, sources, or stakeholders disagree in ways that cannot be dismissed as bad faith or ignorance.

Coverage

0 of 1 conditions grounded · 1 open.

Read the methodologyDownload the trigger-logic data

Review emphasis

Review this draft with the broader epistemic_humility family. It is likely complementary to future candidates on quantified confidence communication, confidence calibration, domain-specific confidence, evidential-bar adjustment, expert-disagreement humility, knowledge-warrant audit, and revision-readiness precommitment. The most important boundary risks are false balance, mechanism-only collapse, and overlap with observational-equivalence or disconfirming-evidence patterns.

Common Mechanisms

6 documented mechanisms across 4 implementation forms.

The grouping reflects forms represented among the mechanisms currently documented for this archetype; an absent form is not necessarily an impossible implementation.

Analysis, Modeling & Optimization · 1 mechanism

  • Discriminating Test Matrix — A grid crossing rival hypotheses against pieces of evidence, scoring each cell for consistency and each row for reliability — so effort goes to the observations that actually separate the rivals rather than to evidence that fits them all.

Assessment, Review & Assurance · 2 mechanisms

  • Counter-Narrative Probe — Builds the single strongest opposing account of the same facts — the version a sharp skeptic would defend — derives what it predicts we should see, and checks, so the leading story has to beat a real challenger instead of a strawman.
  • Red-Team Rival Explanation Review — A structured review in which an assigned adversary must generate and defend rival explanations against the group's favored conclusion, and the team must answer them on the record before confidence is allowed to settle.

Interface, Display & Cue · 2 mechanisms

  • Base-Rate Alternative Prompt — Asks how often the leading explanation actually holds in cases like this — dragging the boring, common alternative that the reference class favors onto the table before a vivid but rare story is accepted.
  • Why Else Could This Be True? Prompt — A single forcing question — 'why else could this be true?' — asked before confidence hardens, that makes the reasoner produce several other explanations fitting the same facts, converting one satisfying story into a field of candidates.

Representation, Specification & Plan · 1 mechanism

  • Differential Diagnosis List — Enumerates a bounded, ranked list of candidate explanations for the same findings — including the unlikely-but-dangerous ones — then rules each in or out as evidence arrives, so the answer is the survivor of a field rather than the first guess.

Compression statement

Alternative-Hypothesis Generation is the pattern of transforming a single favored explanation into a bounded rival-hypothesis set. It does not demand that every alternative be equally likely. Instead, it preserves epistemic humility by asking which plausible explanations, mechanisms, causes, diagnoses, or assumptions could also fit the observed facts, then mapping what evidence would separate them and adjusting confidence accordingly.

Canonical formula: confidence(leading_claim) := warrant(leading_claim | evidence, plausible_rivals, discriminating_tests, unresolved_uncertainty); do not escalate confidence until rival_fit and evidence_discriminability have been reviewed

Abstractions this archetype builds on — directly (a source ingredient) or as a related pattern. Links follow the typed catalog namespace.

Built directly on (1)

  • Epistemic Humility: Calibrating the confidence of one's claims to the actual strength of the evidence and staying open to revision when new information arrives.

Also references 25 related abstractions

Variants

Narrower or domain-specific specializations that share this archetype's core structure. Recognized variants are established; candidate variants are provisional.

Differential-Diagnosis Hypothesis Generation · domain variant · recognized

Generate a structured list of plausible diagnostic alternatives before treating one diagnosis as settled.

  • Distinct from parent: Narrower because the hypothesis set concerns diagnostic explanation of an observed state.
  • Use when: A symptom pattern or observed state has more than one clinically, operationally, or technically plausible cause; Premature closure would cause harm because the first matching diagnosis may be wrong or incomplete.
  • Typical domains: medicine, incident response, machine fault diagnosis
  • Common mechanisms: differential diagnosis list, base rate alternative prompt, discriminating test matrix

Root-Cause Rival Explanation Generation · domain variant · recognized

Generate several plausible causal explanations for a failure, anomaly, or outcome before selecting a root cause.

  • Distinct from parent: Narrower because it focuses on causal attribution after an event or anomaly.
  • Use when: The same failure could result from process, incentive, human, technical, environmental, or data causes; The organization is tempted to blame the most salient actor, team, event, or recent change.
  • Typical domains: postmortems, quality improvement, safety investigations
  • Common mechanisms: five whys branching variant, counter narrative probe, timeline contradiction check

Scientific Rival-Hypothesis Generation · domain variant · recognized

Formulate plausible rival explanations and discriminating predictions before treating a scientific hypothesis as strongly supported.

  • Distinct from parent: Narrower because it uses scientific evidence standards and test design.
  • Use when: Evidence supports a favored model but may also be compatible with confounding, measurement artifact, selection bias, or a competing mechanism; Researchers need to distinguish exploration from confirmatory claim-making.
  • Typical domains: experimental science, observational research, model comparison
  • Common mechanisms: pre registered rival hypothesis table, discriminating test matrix, robustness rival check

Strategic-Assumption Alternative Generation · governance variant · recognized

Generate plausible alternatives to the assumptions supporting a strategy before committing resources.

  • Distinct from parent: Narrower because the alternatives are decision-relevant strategic assumptions rather than all possible explanations.
  • Use when: A plan depends on claims about customers, competitors, regulation, costs, behavior, or timing; Decision makers agree too quickly because the plan story feels coherent or politically convenient.
  • Typical domains: strategy, policy design, product roadmapping
  • Common mechanisms: premortem rival assumption prompt, red team rival review, small bet discrimination probe

Investigative Counter-Narrative Generation · communication variant · candidate

Construct a credible counter-narrative that explains the same evidence before presenting an investigative conclusion.

  • Distinct from parent: Narrower because narrative coherence and communication are central.
  • Use when: A coherent narrative risks hiding ambiguity, missing evidence, or interpretive dependence; The audience may mistake narrative coherence for evidential certainty.
  • Typical domains: journalism, historical analysis, legal investigation
  • Common mechanisms: counter narrative probe, source trace challenge, contradiction inventory

Near names: Alternative Explanation Generation, Rival Hypothesis Generation, Competing Hypotheses Review, Differential Diagnosis Generation, Counter-Narrative Generation.

Editorial Notes

Problem Classification

Classification: Uncertainty, Evidence & Inference Failure → Explanatory Hypothesis, Pattern & Case Reasoning

Problem kernel: a favored explanation closes inquiry before rivals exist

Rationale: Evidence underdetermines the conclusion, yet salience or convenience closes inquiry before alternatives and tests are named.

Independent corroboration: The earliest necessary condition in the frozen evidence is: A reasoner, team, institution, or model converges on a favored explanation before credible alternatives have been named, compared, or tested. That is a explanatory hypothesis pattern and case reasoning problem because Partial or recurring observations are forced into a favored explanation, analogy, pattern, or universal before alternatives and boundary cases are tested.

Review outcome: Independent reviewer agreement; high confidence.