When disagreement persists despite shared facts, draw both mental models side by side
When confusion or disagreement persists despite shared facts, externalize all mental models spatially (whiteboard, diagram, parallel columns) before continuing verbal discussion, because visual comparison reveals structural misalignment that sequential verbal exchange cannot surface.
Why This Is a Rule
When two people agree on the facts but disagree on the conclusion, the disagreement is almost never about logic — it's about structure. They have different mental models: different causal relationships between the same facts, different assumptions about which factors matter most, different implicit hierarchies of priority. Verbal discussion can't resolve this because speech is sequential — you can only present one model at a time, and the listener must hold your model in memory while comparing it to their own. Working memory can't do this reliably.
Spatial externalization solves the problem by making both models simultaneously visible. Two diagrams on a whiteboard, two flowcharts side by side, two parallel columns — whatever format makes the structure visible. The moment both models are visible, the structural divergence becomes obvious: "Oh, you think A causes B, but I think B causes A" or "We agree on everything except whether X or Y has priority."
This type of disagreement can persist for hours in verbal discussion because the participants don't realize they're operating from different structures. They argue about conclusions when the real divergence is in premises.
When This Fires
- Two people have discussed an issue multiple times without resolution despite apparent agreement on facts
- A team debate feels circular — the same arguments keep surfacing without progress
- Someone says "I agree with everything you said, but I reach a different conclusion"
- Any situation where confusion persists despite seeming alignment on information
Common Failure Mode
Drawing one model on the whiteboard and asking "does everyone agree?" This tests whether people can follow your model, not whether their model matches. You need both models visible simultaneously. Ask each participant to draw their understanding independently, then compare. The differences between the drawings are the actual disagreement — which is often not what anyone thought they were disagreeing about.
The Protocol
When discussion stalls despite shared facts: (1) Stop talking. (2) Give each participant a whiteboard section, paper, or digital canvas. (3) Ask everyone to independently draw how they understand the relationship between the key factors — boxes and arrows, flowcharts, Venn diagrams, whatever captures structure. (4) Display all models simultaneously. (5) Walk through the structural differences: "Here's where our models diverge." The conversation shifts from arguing conclusions to comparing structures, which is both faster and more productive.