If testing and interleaving clearly work, why do learners keep abandoning them?

Metacognitive illusions are stubborn — even after directly experiencing that a hard method works better, many learners revert to the easier method that merely feels productive.

You can show someone their own data — they recalled more after self-testing than after re-reading — and a large share will still go back to re-reading. The problem isn't ignorance of the evidence; it's that the feeling of fluency during re-reading is immediate and persuasive, while the payoff of testing is delayed and invisible in the moment.

So knowing the science isn't sufficient. You have to keep deciding against the feeling, repeatedly, because the illusion regenerates every study session.