Blend Coherence Review¶
Review protocol — instantiates Conceptual Blending for Innovation
A critical read-through that hunts a candidate blend for internal contradictions, unsupported transfers, hidden assumptions, and imported risks before it ships.
The Blend Coherence Review is an adversarial read of a blend that already exists on paper: a structured protocol that goes looking for the ways the hybrid secretly doesn't hold together. Its defining stance is critical, not generative — it builds nothing and proposes nothing; it interrogates. Where a canvas states the blend and a prototype tests it in the world, the review examines the stated blend analytically for four specific pathologies: internal contradictions between the source logics, transfers asserted without support, assumptions carried over silently, and risks imported from a source domain into a context that can't safely hold them. It ends not with a verdict of "good" but with a map of tensions, an explicit statement of where the blend must not be extended, and a flag on any imported hazard — the honest diagnostics a decision can act on.
Example¶
A corporate learning team has a blend they love: combine an escape room (immersive, timed, team-based puzzle-solving) with mandatory compliance training (regulatory content, individual accountability, auditable completion) into a "compliance escape room" where teams solve puzzles built from real policy scenarios. Before it's rolled out to twelve thousand employees, it goes through a Blend Coherence Review.
The reviewers work the four pathologies. Internal contradiction: an escape room rewards teamwork and shared success, but compliance training requires individual, auditable proof that each person learned the material — a team that solves the room together leaves no record of who understood what. That's a real tension, mapped, not waved away. Unsupported transfer: the claim that "fun improves retention of policy" is asserted from the escape-room side without evidence for regulatory content specifically. Hidden assumption: the blend assumes failing the room is safe fun, but failing compliance training has legal consequences — the emotional register silently conflicts. Imported risk: escape rooms use time pressure and mild deception, and importing those into training about, say, harassment reporting could trivialize the subject — a hazard flagged for risk review. The output is a tension map plus a limit note ("do not apply escape-room deception or time-pressure to sensitive-conduct content") that lets the team fix or bound the blend before launch, not after a complaint.
How it works¶
The review is a checklist-driven interrogation run by someone other than the blend's authors. First, surface every cross-source tension: for each pair of imported elements, ask where the two source logics pull in opposite directions (teamwork vs. individual audit), and record them rather than resolving them on the spot. Second, test coherence analytically: does the blend contradict itself once its assumptions are made explicit? Third, flag unsupported transfers and hidden assumptions: which claims ride on a source's prestige rather than evidence, and which assumptions crossed silently. Fourth, run a risk review on anything the blend imports into a consequential context, and write the limit note marking where the blend must not be extended. The discipline is authorial distance — the review is worthless if run by the people invested in the blend's success, because they will explain the tensions away.
Tuning parameters¶
- Reviewer independence — authors self-reviewing versus a fresh panel. Greater distance catches more explained-away tensions but costs coordination and can feel adversarial.
- Pathology depth — a quick coherence pass versus a full hunt for contradictions, transfers, assumptions, and risks. Deeper reviews catch subtler failures but slow the pipeline.
- Risk sensitivity — how consequential a context must be before the risk-review step is mandatory. High sensitivity protects safety-critical blends; blanket application burdens playful ones.
- Output form — a pass/fail gate versus a diagnostic tension map. Diagnostics preserve the blend for repair; gates are faster but blunter.
- Stopping rule — reviewing until no new tension appears versus a fixed timebox. Exhaustive review is thorough but has diminishing returns.
When it helps, and when it misleads¶
Its strength is catching the failures that enthusiasm hides: the teamwork-versus-audit contradiction is invisible to a team in love with "compliance escape room," and only a distanced, adversarial read surfaces it before rollout. Its spirit is that of the premortem — deliberately assuming the idea has failed and asking why, to expose weaknesses while they can still be fixed.[n1]
Its failure mode is review as veto theater — a critique so eager to find fault that it kills coherent, useful blends for tensions that are actually resolvable, choking the archetype's generativity. The opposite misuse is the captured review, run by the authors, that rationalizes every tension into a feature. The guarding discipline is to keep the review diagnostic: its job is to name tensions, unsupported transfers, and imported risks precisely enough that the blend can be repaired or bounded — not to render a thumbs-up or thumbs-down, which belongs to the people who own the decision.
How it implements the components¶
This protocol realizes the critical-scrutiny face of the archetype:
tension_or_mismatch_map— its core output is a mapped set of the contradictions between the blend's source logics.coherence_test— it analytically checks whether the blend contradicts itself once its assumptions are made explicit.blend_limit_note— it writes the boundary marking where the blend must not be extended (e.g., no deception in sensitive-conduct content).risk_review— it flags hazards imported from a source domain into a context that cannot safely absorb them.
It does not perform the cross_space_mapping, selective_projection, or blended_space construction that generate the blend — that is Concept Blend Canvas's work — nor does it embody the blend as a prototype_or_probe, which is Hybrid Prototype's; this review only scrutinizes a blend that already exists on paper.
Related¶
- Instantiates: Conceptual Blending for Innovation — supplies the analytic scrutiny that guards a blend against incoherence and imported harm.
- Consumes: Concept Blend Canvas supplies the stated blend the review interrogates.
- Sibling mechanisms: Concept Blend Canvas · Hybrid Prototype · Design Mashup Workshop · Cross-Domain Innovation Sprint · Business Model Pattern Mixing · Interdisciplinary Model Synthesis · Forced Connection Exercise · Metaphorical Blend Prompt
Editorial Notes¶
Form Classification¶
Form family: Assessment, Review & Assurance
Rationale: A critical read-through that hunts a candidate blend for internal contradictions, unsupported transfers, hidden assumptions, and imported risks before it ships, making its operative form a bounded evaluation of existing evidence or work that produces a finding or disposition.
Independent corroboration: The frozen evidence defines Blend Coherence Review as 'A critical read-through that hunts a candidate blend for internal contradictions, unsupported transfers, hidden assumptions, and imported risks before it ships', so its operative form is Assessment, Review & Assurance.
Review outcome: Independent reviewer agreement; high confidence.
Origin Attribution¶
Primary origin: Cognitive Science
Origin pattern: Cross-disciplinary synthesis
Present-day reach: Multi-domain
Rationale: The object under review is a conceptual blend, a construct formalized in cognitive linguistics and cognitive science.
Related originating lineages:
- Engineering & Design — Design critique contributes the pre-release review of transfer assumptions, contradictions, and imported risks.
- Philosophy — Conceptual analysis supplies tests for consistency and unsupported inference.
Review resolution: Cognitive science is the agreed primary lineage through conceptual blending and cross-space mappings. Philosophy contributes coherence and contradiction tests, while engineering contributes adversarial design review; the resulting review is a cross-disciplinary Encyclopedia synthesis.
Encyclopedia synthesis: The exact catalogued form synthesizes established practice rather than reproducing a single standard historical label.
Review outcome: Reconciled after independent review; high confidence.
Notes¶
[n1] A premortem, described by Gary Klein, has a team imagine a plan has already failed and work backwards to the reasons — a prospective-hindsight technique that surfaces weaknesses a team invested in an idea would otherwise rationalize away. The review applies the same distanced, failure-seeking stance to a candidate blend. ↩