Mental Models

The world is too complex to reason about from raw data. A mental model is a simplified internal representation of how something works — a mental map of a domain that lets you predict, decide, and act without solving the full problem each time (Craik, 1943; Johnson-Laird, 1983). Models compress experience. Once you have a working model of gravity, you no longer compute trajectories; you just catch. Once you have a model of a familiar coworker, you no longer analyze every sentence; you interpret. This compression is why models are useful and why they are dangerous. They save effort by discarding information — and sometimes what they discard is exactly what matters.

Every model is a lens: it makes some features visible and hides others. A financial model highlights costs and revenues but rarely surfaces relational damage. A moral model highlights harm but may miss efficiency. Kahneman and Tversky's work on heuristics and biases (1974) documented systematically how the fast, model-driven mode of thinking produces predictable errors when the model does not fit the situation. This is why over-reliance on any single model is dangerous. A person whose only model is economic exchange sees exchange everywhere; a person whose only model is threat sees threats everywhere. Charlie Munger's advice to build a latticework of models across disciplines captures the practical response: many models per situation reduce the blindness of any one.

Across traditions, a handful of models recur because they generalize well: feedback loops from systems thinking (reinforcing and balancing loops shape most complex behavior); opportunity cost from economics (the value of what you gave up to choose what you chose); base rates from probability (the prior probability of a category before evidence); second-order effects (what happens after the first-order response); inversion (asking what would guarantee failure and avoiding those actions); and steel-manning from rhetoric (stating the opposing view in its strongest form before responding). Each model is a lens. Adding lenses does not make the world simpler; it makes your interpretation of the world less brittle.

For any important decision, deliberately walk through at least three models. If you naturally reach for one framing, force yourself to also apply a second and a third. Notice which model your intuition preferred and why. Over time, this practice does two things. It surfaces the invisible assumptions embedded in your default model. And it strengthens the meta-model — the awareness that any conclusion is model-dependent. That awareness is much of what wisdom is.