Judgment of learning
Your on-the-spot prediction of how well you'll later remember something.
What it means
A judgment of learning is a metamemory prediction, made during or just after study, of the likelihood that an item will be remembered on a future test. Such judgments are inferential rather than direct readouts of memory strength, drawing on cues like current retrieval fluency, perceptual ease, and beliefs about learning. They are systematically miscalibrated in revealing ways: delayed judgments are far more accurate than immediate ones, and easy-but-shallow processing breeds overconfidence. These illusions explain why learners often misallocate study time, favoring material that merely feels fluent. The construct is central to research on self-regulated learning and to designing study strategies that counter fluency-based overconfidence.
Examples
Right after rereading a passage it feels mastered, yet a delayed self-test reveals far weaker retention than predicted.
Swiping through vocabulary cards, each word feels known the instant it appears; a week later, tested cold, half are gone — fluency in the moment fooled the prediction.
Watching a colleague's clear tutorial, you feel certain you could repeat every step; the delayed truth arrives when you sit at the keyboard alone and cannot.
First described in Arbuckle & Cuddy (1969); delayed-JOL effect, Nelson & Dunlosky (1991).