Mental Models¶

Definition¶

Chris Argyris's distinction anchors the discipline: people behave consistently with their theory-in-use (what their actions actually imply), not their espoused theory (what they say they believe). Mental models are dangerous not because they're wrong — all models simplify — but because they're tacit: unexamined, they get mistaken for fact, and reality can drift far from them before anyone notices.

In the Book¶

Detroit's Big Three ran for decades on an unexamined assumption set ("styling matters more than quality") that had once been true and was mistaken for permanent — losing US market share to imports (near-zero to 38% by 1986) as a result. The counter-case is Royal Dutch/Shell: planner Pierre Wack discovered in 1972 that handing managers accurate forecasts of coming oil-market discontinuity changed nothing ("our scenarios would be like water on a stone") until scenarios were redesigned to force managers to see which assumptions their own trouble-free future required — after which Shell went from weakest of the seven oil majors (1970) to arguably strongest (1979). Two techniques operationalize the discipline: the left-hand column (writing what you're really thinking beside what you actually said, to expose untested assumptions steering a conversation) and balancing inquiry and advocacy (stating your reasoning while genuinely probing others', replacing "winning the argument" with "finding the best argument").

Why It Matters¶

Good ideas fail in practice not from weak will but because they collide with mental models nobody has surfaced. Senge's closing metaphor: systems thinking without mental models is the DC-3's engine without wing flaps — necessary, insufficient.

Connections¶

parallels value-loading — both name the same underlying danger: implicit, unexamined internal models (a manager's theory-in-use; an AI system's actual learned values) drive behavior far more than any explicitly stated goal or "espoused theory," and the gap surfaces only under scrutiny most systems never receive.
extends team-learning — Ch. 12's dialogue practice (suspending assumptions collectively) is Ch. 10's individual reflection and inquiry skills applied at the level of a whole team.