CONCEPT
The Questioning Muscle
Tetlock's metaphor for the trainable, loseable cognitive capacity to evaluate claims rather than absorb them—the habit of asking ‘how confident am I, and how confident should I be?’ that AI output’s confident fluency is most likely to erode.
The questioning muscle is not a metaphor. It is a measurable cognitive capacity that, as Philip Tetlock’s research on
superforecasting demonstrated, improves with structured practice and degrades without it. The
Good Judgment Project’s training protocol showed that an hour of instruction in probabilistic reasoning produced lasting improvements in forecast accuracy—not because it taught new facts but because it installed new habits: the habit of assigning a probability to confidence, checking it against alternative sources, and updating when the evidence says you were wrong. The analogy to physical fitness is not decorative: regular exercise against resistance produces growth; detraining produces atrophy; and the atrophy can proceed for a long time before it becomes visible, because the person losing the capacity does not feel the loss in real time. The AI environment presents a specific and measurable risk to this capacity through three mechanisms: the elimination of evaluative friction (AI output arrives complete and polished, requiring no engagement