The Same Hours, Rearranged
The same total study time produces far more durable memory when distributed across sessions separated by days than when massed into one sitting — and the best gap between reviews scales to roughly 10 to 20 percent of how long you need to remember. · 11 min
Here is a result that sounds too generous to be real: you can remember more without studying a single extra minute. Take the hours you were going to spend and rearrange them. Four hours in one sitting the night before a test, against the same four hours split across four evenings over two weeks — same material, same total time, same person. Measured a month later, the split-up schedule wins, and not by a little. Packing study into one unbroken block is what most people mean by cramming. Spreading the same work across separated sessions is its rival. The gaps between sessions turn out to be doing quiet, measurable work, and this folio is about how much.
Guess before you learn
You have four hours to prepare for a test three weeks away. Plan A: all four hours tonight, in one focused block. Plan B: one hour tonight, then one hour every few days across the three weeks. Same material, same four hours. On the test, which plan leaves you remembering more?
Plan B, comfortably. The finding is called the spacing effect, and it is among the most reliable results in the study of memory — Cepeda and colleagues' 2006 review pooled over 250 experiments and found distributed practice beat massed practice again and again. If you chose Plan A, you chose the way studying feels: one long block feels productive and leaves you confident. That confidence is folio 4's fluency illusion, and a test three weeks out does not honor it.
9–12
3–5
Suppose you have twenty spelling words. You could practice all twenty tonight until you know them cold. Or you could practice some tonight, some tomorrow, and some the day after that.
The spread-out way works better, even though it takes the same total time. Coming back to the words after a gap — a day, then two — is what makes them stick. Doing them all at once feels faster, but you forget more of them by the test.
6–8
Massed practice packs all your study of a topic into one unbroken session — cramming. Distributed practice spreads the same total time across sessions separated by hours or days. Hold the hours fixed and change only the spacing, and the distributed schedule produces more durable memory nearly every time. This is the spacing effect, one of the most reliable findings in over a century of memory research.
Why does a gap help? Each time you return after forgetting a little, you have to work to bring the material back — and that effortful return is folio 5's retrieval, the very act that strengthens memory. Massed practice never lets you forget enough for the return to be worth anything.
9–12
The size of the gap matters, and it is not fixed. Cepeda, Vul, Rohrer, Wixted, and Pashler (2008) had more than a thousand people learn facts, then varied both the gap between two study sessions and the delay until the final test. The best gap grew with the delay: the longer you need to remember something, the wider the ideal spacing between reviews.
Their rule of thumb: space reviews at roughly 10 to 20 percent of the time you need to hold the material. For a test in a week, that is about a day between sessions; for a test in a year, a few weeks. Space too little and you have merely massed; space too much and you forget everything before the return.
K–2
You have ten new words to learn. You could practice all ten right now, over and over, until bedtime. Or you could practice a few today, a few tomorrow, and a few the day after.
Spreading practice across days makes the words stay longer. Small bits, on different days, beat one big pile all at once — even though it is the same amount of practice.
Undergrad
The spacing effect resists the usual moderators. It appears across verbal materials, motor skills, and classroom subjects, and in Cepeda and colleagues' (2006) synthesis of 254 studies the distributed advantage was near-universal. The 2008 ridgeline study added the crucial refinement: final retention is a non-monotonic function of the gap, and the optimal gap, expressed as a proportion of the retention interval, shrinks as that interval lengthens.
Bahrick and Phelps (1987) supplied the durability case. They taught Spanish vocabulary at intersession gaps of 0, 1, or 30 days; eight years later the 30-day group retained more than twice what the massed group did. The mechanism is argued between study-phase retrieval and encoding variability, but the applied verdict is settled: spacing is one of the few genuinely cost-free gains in study.
Postgrad
Two mechanistic accounts dominate. Study-phase retrieval holds that a spaced repetition triggers retrieval of the first encoding, and that retrieval — more effortful after a gap — does the strengthening. Encoding variability holds that separated sessions occur in more varied contexts, laying down more retrieval routes. The accounts are not exclusive, and a deficient-processing view adds that massed repetitions are simply attended less.
The nonlinearity is the design-relevant part. Because the optimal-gap-to-retention ratio declines with the retention interval, a single fixed gap cannot be optimal across horizons — which is precisely the problem adaptive schedulers solve by expanding intervals after each success. Folio 8 opens one such scheduler and traces its arithmetic on a single card.
spacing effect
The finding that the same total study time produces more durable memory when distributed across separated sessions than when massed into one. Also called distributed practice.
Why is this true?
Why does leaving a gap between study sessions strengthen memory more than studying with no gap?
Because a gap lets you forget a little, so returning to the material forces an effortful retrieval instead of a passive reread. Massed repetition, with nothing forgotten in between, gives retrieval nothing to do — the words are still sitting in mind from a moment ago.
Read the shape, because both ends of it cost you. A gap of zero is massing under another name, and it loses. A gap far longer than the delay to the test is worse still: the first session has mostly vanished before the second arrives, so there is nothing left to reinforce. Between those two failures sits a broad peak. You do not have to land on it exactly — the top of the ridge is nearly flat — but you do have to leave the massing zone, and almost everyone starts there.
To put the effect to work, stop asking how many hours and start asking where they land. Fix your total study time, then break it into shorter sessions on different days. Set the first gap short: the forgetting curve is steepest on day one, so an early review earns the most (folio 3). Then widen each following gap as the material proves it can hold. A working default is to space reviews at about 10 to 20 percent of the horizon you are studying for. The University already does this for you — every folio you finish returns in the Fading Ink, review what's fading, on gaps that stretch wider as your memory holds.
Plan spaced reviews for a test 40 days away — the steps fade as you master them
0.10 × 40 = 4 days
0.20 × 40 = 8 days
study day 0; review at day 4, then day 12, then day 26
The spacing effect tells you that gaps help and roughly how wide to make them. It does not tell you the exact day to review each of a hundred different facts, every one fading at its own rate. Doing that by hand is impossible; doing it by rule is not. The next folio opens the algorithm that schedules every card in this Archive — SM-2 — and traces its arithmetic on a single card, step by step, so you can see the machinery that has been quietly spacing your reviews all along.
Note
The Fading Ink does not use one gap for everything. It widens the gap for each card on its own, based on how well you recall it — the machinery folio 8 lays bare.
Practice — new ink and old, interleaved
1.Which study plan is distributed rather than massed?
2.From folio 5: a gap makes the second study session more valuable. Why exactly?
3.Which pair earned the top utility rating in Dunlosky and colleagues' 2013 review?
4.From folio 3: where is the forgetting curve steepest, and what does that imply for timing your first review?
5.From folio 4: cramming feels more productive than spacing. What is that feeling actually reading?
6.Match each term to its meaning.
7.By the 10-to-20-percent rule, for a test 50 days away, a gap near the middle of the range is about how many days?
8.Turn this highlighted sentence into a self-test question: 'The hippocampus replays the day's learning to the cortex during slow-wave sleep.'
9.From folio 6: you open a spaced review session and find you have forgotten several items. Is that forgetting a problem?
10.From memory: for a test in one year, is the ideal gap between reviews larger or smaller than for a test in one week, and by roughly how much?
Larger. The ideal gap grows with the retention interval — about a day for a one-week test, but a few weeks for a one-year test, following the 10-to-20-percent rule.
How close were you? Grade yourself honestly — it sets your review date.
11.Without looking back: what do judgments of learning mostly ride on, and why does rereading inflate them?
They ride on how fluently the material processes right now; rereading raises fluency without adding much to memory, so the judgment climbs while recall stays flat.
How close were you? Grade yourself honestly — it sets your review date.
12.By the figure above, a twelve-digit list comes back with about how many digits correct?
13.Which change most improves the accuracy of your judgments of learning?
14.Original learning took 25 minutes; relearning a month later took 20. What is the savings, in percent?