Skip to content

Concept Blend Canvas

Template — instantiates Conceptual Blending for Innovation

A single worksheet that names the two source spaces side by side and captures their projection into one blended concept, so nothing in the blend stays implicit.

The Concept Blend Canvas is a fixed-layout worksheet — printed poster, whiteboard template, or shared doc — with a cell for every part of a blend: a panel for Input Space A, a panel for Input Space B, a middle column for the correspondences between them, a shortlist of the elements actually carried across, and a central area where the hybrid concept is drawn as a single thing. Its whole job is to make the blend legible on one surface: what came from where, what was mapped to what, what was deliberately left behind, and what the combined idea now is. What makes it this mechanism rather than a combinatorial grid like the Morphological Matrix is that the canvas holds two named, richly-described conceptual spaces and traces their projection into one blend — it is not a lattice of independent parameters whose cells you enumerate. The canvas doesn't decide whether a blend is any good; it forces the blend to be stated completely enough that judging it becomes possible.

Example

A city library is stuck: circulation is falling and the board keeps proposing variations on "quieter rooms, newer books." A staff team pins up a Concept Blend Canvas to escape the rut. In Input Space A they describe the library structurally — free access, a professional curator, silence, borrowing-and-returning, a reference desk. In Input Space B they describe an independent bookstore — browsing by mood, a café, staff picks, events, buying to keep. The middle column maps the correspondences: curator ↔ bookseller, reference desk ↔ staff-pick table, borrow-return ↔ buy-keep. The projection shortlist keeps the bookstore's browsing-by-mood displays, café, and events and the library's free access and curator, and pointedly leaves behind both silence and the cash register.

Drawn in the center, the blend becomes a "living-room library": mood-curated display tables, a coffee bar, evening author events, everything still free to borrow. Because every panel is filled in on one wall, the board can see exactly which library values survived and which bookstore mechanics were imported — and the fact that the till was left out is a visible decision, not an omission. That completed canvas is what the next steps (a coherence review, a small pilot) act on.

How it works

The canvas enforces a filling order rather than a free sketch. First, both input spaces are described structurally, not by label — the panels demand functions, roles, and constraints, so "a bookstore" becomes a set of mechanics the blend can actually draw on. Second, the mapping column is filled before the center — you must find the correspondences (this-plays-the-role-of-that) before you're allowed to draw the hybrid, which blocks the jump straight to a clever name. Third, the projection shortlist is explicit and subtractive: every carried-over element is written down, and the empty space around it is the record of what was excluded. Only then is the central blend drawn as one integrated concept with its own vocabulary. The canvas's discipline is completeness: an empty panel is a visible hole, so half-thought blends can't hide.

Tuning parameters

  • Panel granularity — coarse panels (one box per space) or fine ones (separate boxes for roles, constraints, incentives). Finer panels surface more mapping opportunities but slow the fill and can fragment the idea.
  • Projection explicitness — whether the "left behind" list is a required cell or optional. Making exclusion mandatory catches one-source-domination early but adds friction.
  • Number of spaces — a two-space canvas stays legible; adding a third panel enriches the blend but the mapping column grows quadratically and the surface gets crowded.
  • Fidelity of the center — a one-line label, a sketch, or a fully specified concept. A richer center is more testable but tempts the team to over-commit before review.
  • Reuse stance — a throwaway per session or a filed artifact kept across cycles; kept canvases become a searchable library of attempted blends.

When it helps, and when it misleads

Its strength is legibility: by giving every part of a blend a home on one surface, the canvas makes the difference between a real blend and a slogan obvious — a "living-room library" with an empty mapping column is exposed as a name, not a concept. It is the working surface for what Gilles Fauconnier and Mark Turner called conceptual blending: two mental spaces, a partial mapping between them, and a projection into a third.[n1]

Its failure mode is template theater — filling every cell produces the feeling of rigor without the substance, and a canvas can be neatly completed around a blend that is incoherent or useless. The classic misuse is treating a full canvas as a finished decision, when all it has done is state the blend; it runs no coherence or usefulness test of its own. The guarding discipline is to treat a completed canvas as an input to review and prototyping, never as their verdict, and to read the empty space around the projection shortlist as carefully as the boxes that are filled.

How it implements the components

The canvas realizes the capture-and-structure face of the archetype — the components that must be made explicit before anything can be tested:

  • input_space_a — its first panel forces a structural description of the first source domain, not just a name.
  • input_space_b — its second panel does the same for the second source, side by side for comparison.
  • cross_space_mapping — the middle column records the correspondences (role-plays-role) that make the projection principled rather than loose.
  • selective_projection — the projection shortlist names exactly which elements cross and, by omission, which are left behind.
  • blended_space — the central area is where the projected elements are drawn as one integrated hybrid with its own logic.

The canvas does not run the coherence_test or usefulness_test — those are performed by Blend Coherence Review and Hybrid Prototype; the canvas only lays the blend out completely enough for them to judge it.

Editorial Notes

Form Classification

Form family: Representation, Specification & Plan

Rationale: A single worksheet externalizes two source spaces, their projected elements, and the resulting blended concept in named fields, so its operative form is a structured concept representation.

Nearest alternative: Interface, Display & Cue — The canvas guides what the author enters, but the practitioner retains and reviews the completed blend artifact rather than relying on an interactive user-facing control.

Review outcome: Adjudicated after independent review; high confidence.

Origin Attribution

Primary origin: Cognitive Science

Origin pattern: Cross-disciplinary synthesis

Present-day reach: Multi-domain

Rationale: Conceptual-blending theory cohered input mental spaces, cross-space mappings, selective projection, and emergent blended structure.

Related originating lineages:

  • Engineering & Design — Canvas and worksheet practice turns the mental-space model into a visible collaborative design artifact.
  • Linguistics & Semiotics — Cognitive linguistics developed conceptual integration as a method for explaining meaning construction across input spaces.

Review resolution: Conceptual-blending theory in cognitive science supplies input spaces, mappings, selective projection, and emergent structure. Cognitive linguistics is a formative home of that theory, while design practice supplies the collaborative canvas form, making the Encyclopedia worksheet a cross-disciplinary 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

The Concept Blend Canvas is the near-twin of the already-authored Morphological Matrix, and the two are easy to confuse because both are structured surfaces. The line between them: the canvas maps and projects two named conceptual spaces into one blend, whereas the matrix enumerates combinations of a single problem's independent parameters. If your surface has two rich source panels and a mapping column, it's a canvas; if it's a grid of parameter columns you read across, it's a matrix.

[n1] In Fauconnier and Turner's conceptual blending (or "conceptual integration") theory, meaning is often built by taking two input mental spaces, establishing a partial cross-space mapping, and projecting selected structure into a third blended space that can develop emergent structure of its own. The canvas is a practical layout of exactly those parts.