How can you tell whether you really know something?

You can tell whether you know something thanks to metacognition, your mind's ability to watch itself, or, in plain words, thinking about your own thinking. The psychologist John Flavell gave it its name in 1979. You use it when you notice that one subject is harder for you than another, or when something makes you think, "I should double-check that."

Psychologists split it into two jobs. Monitoring is the judgment: do I know this? Control is what you do with that judgment: keep studying, move on, look it up. When students revise, they constantly make these "judgments of learning", and those guesses decide where their study time goes.

The tip-of-the-tongue feeling is metacognition you can feel. In 1966, Harvard researchers Roger Brown and David McNeill read people definitions of rare words. When someone was sure the word was "right there", they could often name its first letter or count its syllables, better than chance. The feeling of knowing wasn't an illusion.

But the monitor can lie. Students often mistake a lack of effort for understanding, and the most confident judgments tend to be the least accurate. That's why material that went down easily can feel learned when it isn't. The good news is that it can be trained: students taught to pretest themselves, evaluate their work and plan their study did better on exams.