Spaced Repetition: Why the 1-3-7-16-35 Ladder Works

Re-reading a chapter feels productive and mostly is not: the words look familiar, and familiarity masquerades as memory. Spaced repetition replaces that illusion with a schedule — review material in expanding intervals, timed to land just as forgetting begins. The classic ladder runs 1, 3, 7, 16 and 35 days, and its shape is not arbitrary.

The forgetting curve it fights

Hermann Ebbinghaus measured it in the 1880s and every replication since has agreed on the shape: memory decays steeply at first and then flattens. Within a day, most of what you learned has faded; what survives the first days survives much longer. The curve is exponential-ish, which is why the review intervals that counter it expand geometrically rather than staying fixed — each review has to reach further into the future than the last.

Reviews placed too close together are almost wasted: you are rehearsing material you have not yet had time to forget, which feels easy and builds little. Reviews placed too far apart cost full relearning. The efficient frontier runs between those failure modes — each review scheduled right where the curve starts to drop.

Why 1, 3, 7, 16, 35?

Look at the gaps between the classic intervals: 2, 4, 9, 19 days. Each gap roughly doubles, mirroring the curve's flattening. The sequence also fits real study rhythms — the first review lands the next morning (consolidating overnight), the second at the three-day mark (the first dangerous drop), the third at a week, and the last two stretch across a fortnight and a month, converting crammed material into knowledge that survives a semester.

Surviving the 35-day review is the graduation test: if you can still recall it a month later, the item has effectively moved into long-term storage and only needs occasional spot-checks. Five successful reviews is, in practice, what it takes most items to get there.

What a "successful review" means — and what failure does

The ladder only works if each review is a genuine retrieval attempt: answer first, check after. Recognising the material as it is re-read does not count; forcing recall before looking is the entire mechanism. When an item fails its review, the honest response is to reset it to the start of the ladder — a failed recall means the forgetting curve restarted, and scheduling that item 16 days out would be planning to review a memory that no longer exists.

Difficulty is information. Items that keep failing early reviews are badly understood, not badly memorised — they need a different explanation, not more repetitions of the same one.

Fixed ladders versus adaptive software

Modern SRS apps (Anki and relatives) adjust each item's next interval based on how hard the recall felt, using algorithms descended from SuperMemo. That adaptivity is genuinely useful at scale — thousands of cards, months of history. But a fixed ladder has virtues of its own: it is transparent, printable, shareable and explainable to a study group, and it needs no app at all. For an exam block with 30 topics and three weeks, a fixed 1-3-7-16-35 plan computed once is often followed more faithfully than an algorithm nobody can see.

The two approaches also answer different questions. Adaptive systems optimise long-term retention per minute of review; a fixed ladder optimises "will I know this by the exam on the 14th?" — a date-anchored question that fits a planner better than a scheduler.

Running the plan in practice

Keep one list: item, stage (how many reviews it has survived), and the date of the last review. The next date is mechanical — add ladder[stage] days. Each day, do everything the schedule flags as due or overdue before learning anything new; reviews beat new material when time is short, because protecting an old investment is cheaper than rebuilding one. When you review, bump the stage and move the date to today. That is the entire system — the discipline is not in the math but in never letting today's list slide.

Try it yourself

The Spaced Repetition Planner turns that list into a dated schedule: enter each topic with its stage and last-review date, and get next-review dates, overdue flags and a day-by-day due list on the 1-3-7-16-35 ladder. Protect the review sessions themselves with the Pomodoro Timer or Focus Timer, and check how many days remain before the exam with the Date Countdown.