Spaced repetition counteracts the forgetting curve by timing reviews to land just before forgetting would occur. Each successful recall pushes the next review further into the future, gradually flattening the curve. Distributed practice - spreading study across many days - beats massed practice (cramming) by wide margins. Within spacing, an expanding schedule (e.g., reviews after roughly 1, 3, 7, 14, and 30 days) consistently outperforms a uniform schedule with the same total time invested. The "lag effect" captures a related finding: for the same end-of-test retention, an initial study-test gap of days produces better long-term memory than a study-test gap of minutes, provided each session ends with successful retrieval rather than pure re-exposure.
Retrieval practice, also called the testing effect, is the single most robust learning technique in the experimental literature. Actively pulling information from memory strengthens the retrieval path more than passively seeing the material again. The effect is large: in classic studies, a single short quiz after a lecture roughly doubled retention after a week compared with re-reading. Pre-testing - quizzing on material before it has been studied - primes the brain to encode the upcoming information more deeply, even when the pre-test answers are wrong. Post-testing, a quick quiz shortly after study, produces significantly better delayed retention than re-reading the same material. Closed-book self-tests dominate open-book ones for retention; open-book tests help only when they still require synthesis rather than look-up.
Interleaving mixes different topics or problem types within a single study session rather than blocking on one. It feels harder and lowers in-session performance, but it produces better long-term retention and transfer. The contextual interference effect captures the same idea in motor learning: frequent switching between skills hurts practice performance yet improves long-term retention. Beyond spacing and retrieval, elaborative interrogation - asking explicit "why" and "how" questions and connecting new material to existing knowledge - produces deeper understanding than passive reading. The generation effect shows that attempting to produce an answer, even wrongly, beats simply reading the answer; the struggle to retrieve is itself the learning event.
Two further strategies handle different content well. Dual coding pairs verbal information with visual representations such as diagrams, mind maps, or charts, encoding the material through two channels for stronger recall. Concrete examples, attached to otherwise abstract principles, are stored and retrieved by the brain more readily than the abstractions themselves. Self-reference - asking "how does this apply to me?" - is one of the most reliable encoding boosts known. Together these techniques share a common virtue: each generates effortful processing that produces durable memory. The principle of desirable difficulty sums this up - learning conditions that feel harder in the moment, such as retrieval practice, interleaving, spacing, and generation, typically produce stronger long-term memory than easy conditions, even though the easier ones feel more productive. The AGES model condenses the neuroscience view: Attention, Generation, Emotion, and Spacing act together to drive learning. As a practical rule of thumb, the 85% rule suggests aiming material at roughly 85% accuracy - challenging enough to feel difficult but not so hard that errors dominate.