Mix review of older material with study of new material in every session

Reserve part of every session for retrieving older material before it falls off the forgetting curve entirely.

Why it works

Without deliberate scheduling, older material is naturally crowded out by new content — a form of passive forgetting by displacement. Interleaving old and new material in every session forces each older item to remain active in the retrieval system, preventing the forgetting curve from running to completion. The retrieval of old material also provides schema hooks for the new material, making new encoding more efficient through integration with prior knowledge.

How to do it

1. Begin each learning session with a 5-minute review of material from 3–7 days ago.
2. After studying new material, end with a brief recall of the older items again.
3. If you maintain a question bank, include questions from all previous sessions, not just the last one.
4. Adjust the ratio of old-to-new based on upcoming deadlines and the age of the oldest reviewed material.

Evidence

Interleaving old and new material is supported by both the spacing effect literature and the contextual interference effect. It is standard practice in well-designed spaced repetition systems (Anki, Leitner box). Direct controlled trials for this specific "every-session" protocol are limited. (mechanistic)

The optimal ratio of old-to-new material in a session depends on the review load from prior material; an expanding archive of past material eventually consumes too much session time if not carefully managed.

Common mistake

Studying only new material and scheduling older material for separate "review sessions" that never happen, allowing the forgetting curve to run to completion between sessions.