Augustine of Hippo spent years as a professional rhetorician before becoming a philosopher — and the skill he never fully shed was arguing brilliantly for positions he no longer believed. In the Confessions, he describes this as a kind of inner captivity: the mind can maintain the architecture of an old worldview long after the soul has moved on, running the familiar arguments by reflex. Cognitive scientist Philip Johnson-Laird identified the same mechanism from a different direction: his research on mental models showed that we don't reason from evidence to conclusions, we reason from models to conclusions, and then selectively attend to evidence that fits. The models come first. What Augustine and Johnson-Laird together expose is something more uncomfortable than bias — it's that a mental model doesn't announce when it's become a ghost. It keeps generating arguments, predictions, and explanations with full confidence, even when the conditions that made it accurate have long since changed. The practical move today isn't to doubt everything. It's to pick one framework you use constantly — about your energy, your work capacity, your relationships — and ask when you last genuinely updated it versus when you last successfully defended it.

What is the most confident-sounding explanation you regularly offer — about your own performance, health, or behavior — that you have never once tried to falsify with a concrete test?

Drawing from Cognitive Psychology / Augustinian Philosophy of Mind — Philip Johnson-Laird (Mental Models, 1983), synthesized with Augustine of Hippo (Confessions, c. 397–400 CE)