What is chain of thought reasoning?

Chain of thought reasoning is a concept in personal epistemology: Writing out the steps of your thinking exposes gaps invisible from inside your head. Internal reasoning feels continuous — externalized reasoning reveals the jumps, the missing warrants, the unstated assumptions. The reasoning chain you think you have is not the reasoning chain you actually have until you write it down.

Example: A product manager is deciding whether to delay a launch by two weeks to fix a reliability issue. Inside her head, the reasoning feels airtight: the bug affects 12% of users, the fix is straightforward, and shipping a broken product will damage trust. She is confident in the decision. Then her engineering lead asks her to write a one-page decision memo before the all-hands. She opens a document and begins: 'We should delay the launch because the bug affects 12% of users.' She writes the next step: 'A 12% failure rate is unacceptable for a v1 product.' She pauses. Why is 12% unacceptable? What is her threshold? She does not have one — she had a feeling of unacceptability, not a standard. She writes the next link: 'The fix is straightforward and can be completed in two weeks.' She pauses again. Who said two weeks? The engineer who estimated it, or the engineer who would actually do the work? She has not confirmed. She writes: 'Shipping a broken product will damage trust.' But with whom? The 12% who hit the bug, or all users? And how does she know — has she seen data on trust erosion from comparable bugs in comparable products? Three sentences into her reasoning chain, she has found three gaps: a missing threshold, an unverified estimate, and an unsupported causal claim. Her conclusion may still be correct. But her reasoning was not what she thought it was until she wrote it down.

This concept is part of Phase 10 (Externalization Mastery) in the How to Think curriculum, which builds the epistemic infrastructure for externalization mastery.