Active recall vs re-reading — does the evidence actually say re-reading is useless?

The influencers are directionally right and slightly overclaiming. What the research actually shows: The landmark comparisons (Roediger & Karpicke's testing-effect studies, and the big Dunlosky 2013 review of learning techniques) consistently find practice testing and spaced practice rated high-utility, while re-reading and highlighting rate LOW-utility — not zero. Re-reading produces real short-term familiarity gains; they just decay fast and, worse, inflate confidence more than competence. You feel fluent because the text looks familiar — the classic 'fluency illusion'. Typical effect sizes in those studies aren't subtle: recall-based practice roughly doubling week-later retention versus additional re-reading is a representative result, not a cherry-pick. The honest synthesis: first read — obviously necessary. Second exposure onward, every re-read is a worse investment than the same minutes spent testing yourself on it. Re-reading isn't poison; it's just consistently outcompeted. Migration path for a lifelong re-reader (I was one): after each section, close the book and write 3 questions about it. Next study session, START by answering old questions before any reading. That single habit converts your existing system into recall-based studying without abandoning it.