How Studying Works | Spacing Schedule Shape — Why Expanding Gaps Are Not Automatically Better Than Equal Spacing

HSW-0257 · How Studying Works

A student learns twenty chemistry definitions on Monday.

Should the next reviews happen one day later, then three days later, then a week later? Or should the reviews simply arrive every four days? Or should the first return be later and the final return closer to the examination?

All three schedules can be called “spaced practice”. They are not the same schedule.

The common study slogan—keep making the gaps bigger—sounds scientific because expanding retrieval has a long history in memory research. But the evidence is more conditional than the slogan.

The important variable is not merely whether the gaps expand. It is whether the learner receives enough total spacing, succeeds often enough to retrieve, and is preparing for the retention interval that actually matters.

This article owns a narrow job: the shape of a spaced-retrieval schedule after spacing has already been chosen as the general method. The broader mechanism remains with How Spaced Practice Works. The question here is what changes when repeated reviews are arranged as expanding, equal or contracting intervals.

Quick Answer

Expanding gaps can be useful because early returns keep retrieval successful while later returns introduce more delay. Yet studies comparing expanding schedules with equal spacing frequently find similar final retention once total spacing is controlled. Some experiments find advantages for expanding schedules under particular interference conditions; others find equal spacing equivalent or even better at certain delays. Research also shows that the best spacing depends strongly on how long the knowledge must survive before the final test.

So the better rule is not “always expand”. It is:

Space enough to create retrieval work, but not so much that retrieval repeatedly collapses; then make the schedule serve the retention horizon.

1. Three Ways to Space the Same Number of Reviews

Suppose a learner has three review opportunities after initial study.

• Expanding: 1 day → 3 days → 7 days.
• Equal: 4 days → 4 days → 4 days.
• Contracting: 7 days → 3 days → 1 day.

These labels describe the relative pattern of gaps. They do not tell us the total amount of spacing, whether retrieval was successful at each return, whether feedback was given, what was learned, or when the final test occurred.

That is why two studies can both compare “expanding versus equal spacing” and still produce different results. The actual intervals, material, interference, feedback and final retention test may differ substantially.

2. Why Expanding Spacing Became Attractive

The logic is elegant.

1. Return soon while the memory is still retrievable.
2. Retrieve successfully.
3. Wait longer.
4. Retrieve again under greater difficulty.
5. Keep widening the gap as the memory becomes more stable.

This seems to balance two goals that normally conflict: successful retrieval and increasing retrieval difficulty.

But a plausible mechanism is not the same as a universal scheduling law.

3. What the Comparative Evidence Actually Says

A 2011 experiment on repeated retrieval found a strong effect of absolute spacing: longer total gaps improved long-term retention substantially compared with massed retrieval. Yet the investigators found no inherent final-retention advantage for expanding, equal or contracting relative schedules once spacing was considered. See Karpicke and Bauernschmidt, 2011.

Similarly, research comparing expanding and equally spaced retrieval of text materials found repeated testing beneficial, especially with feedback, but no reliable difference between the two schedule shapes one week later. See Karpicke and Roediger, 2010.

Another long-term vocabulary experiment found equal final recall eight weeks after training for expanding and equal-interval practice, while expanding practice produced higher average accessibility during the weeks of training. See Kang and colleagues, 2014.

That distinction is practical. A schedule can be better at keeping knowledge accessible throughout a course without producing a higher final score months later.

4. The Final Test Date Changes the Answer

The spacing literature repeatedly shows that optimal review distance depends on the final retention interval.

A large quantitative synthesis of distributed practice found that the interstudy interval producing the best later memory tends to increase as the desired retention interval increases. See Cepeda and colleagues, 2006.

An experiment comparing contracting, equal and expanding schedules likewise found that the better schedule changed with the final delay: contracting practice was advantageous at shorter retention intervals in that study, whereas equal and expanding schedules performed better at the 35-day delay. See Küpper-Tetzel, Kapler and Wiseheart, 2014.

This is the first major repair to the slogan:

There is no sensible spacing schedule without a target retention horizon.

5. Retrieval Success Is a Constraint

A gap that is theoretically “desirable” but causes the learner to fail almost every retrieval can become inefficient, especially if feedback is weak or absent.

Expanding schedules partly solve this by making the first return early. The learner gets a successful retrieval while the trace is still accessible, then later gaps can become longer.

But even here the correct conclusion is not that expansion is automatically best. The schedule should be responsive to actual performance.

• If retrieval is effortless every time, the gaps may be too short.
• If retrieval fails repeatedly, the gaps may be too long, the original learning too weak, or the cue too poor.
• If retrieval succeeds only with strong hints, the learner may be learning the cue rather than building flexible access.

6. Retrieval Difficulty Is Not the Same as Calendar Distance

Students often treat days-between-reviews as the difficulty variable.

But retrieval difficulty also changes with:

• how similar competing material is;
• whether the cue is familiar;
• whether context changed;
• whether the learner has used the knowledge elsewhere;
• how well the material was initially understood;
• whether sleep, interruption or new learning intervened;
• whether the final task is recognition, recall, explanation or application.

Two seven-day gaps can produce completely different retrieval difficulty.

7. The Difference Between Maintaining Access and Maximising Final Retention

A school learner rarely cares about one final test only.

The knowledge may need to remain usable during lessons, homework, quizzes and later chapters before the major examination arrives.

This creates two objectives:

• maintenance objective: keep knowledge accessible throughout the learning period;
• terminal objective: maximise retrieval at a specified future date.

An expanding schedule may sometimes be attractive for maintenance because early reviews reduce catastrophic forgetting while later reviews lengthen. Equal spacing may be simpler and may produce similar final retention in many conditions.

The “best” schedule depends on which objective matters.

8. Mathematics Example: Formula Access Across a Term

Suppose a Secondary student learns the quadratic formula in Week 1 but will need it repeatedly until Week 10.

An expanding schedule might look like Day 1, Day 3, Day 7, Day 14, then Day 28. But if the formula appears naturally in mixed homework during the interval, those applications are already retrieval opportunities.

Blindly adding an app-generated review on top of authentic problem use may overschedule the item.

The true schedule is not the flashcard calendar. It is the entire history of successful retrieval.

9. English Example: Vocabulary Recognition Is Not Productive Use

A vocabulary item may be easy to recognise after a week yet still difficult to produce accurately in writing.

One 2026 vocabulary study on contracting, equal and expanding schedules found different patterns depending on receptive and productive knowledge. eduKateSG already treats that specialist vocabulary result separately in Contracting or Expanding Spacing in Vocabulary Learning?.

The broader lesson is important: spacing should be attached to the capability being tested, not merely to the word or topic label.

10. Science Example: Mechanisms Need More Than Definition Recall

A student can retrieve “evaporation is a change from liquid to gas” at every scheduled review and still fail to explain why evaporation causes cooling.

A good schedule therefore changes the retrieval job across returns:

1. definition;
2. diagram;
3. mechanism explanation;
4. prediction;
5. unfamiliar application;
6. mixed discrimination against boiling.

The calendar supports learning. The retrieval content determines what capability is actually maintained.

11. A Practical Spacing Controller

Instead of committing permanently to one shape, use a simple controller.

1. Choose the retention horizon. Tomorrow, two weeks, three months and next year are different goals.
2. Set an initial return soon enough to diagnose the memory.
3. Retrieve before looking.
4. If retrieval is accurate and reasonably independent, widen the next gap.
5. If retrieval fails badly, repair first rather than merely shrinking the next interval.
6. Mix cues and applications as the knowledge stabilises.
7. Keep one delayed check near the actual performance horizon.

This produces something that may look expanding for one topic, equal for another and irregular for a third.

12. Why Algorithmic Spaced-Repetition Apps Still Need Judgment

Spaced-repetition software can estimate when an item should return based on prior responses. That is useful infrastructure.

But the software usually sees a simplified signal: correct, incorrect, easy, hard, perhaps response time.

It may not know whether:

• the answer was guessed;
• the cue was too revealing;
• the learner could explain the idea;
• the knowledge transfers;
• the examination requires another response format;
• a recent lesson already provided retrieval.

Use the scheduler as a reminder engine, not as proof that the capability is mastered.

13. The Most Common Scheduling Errors

• Expanding by rule regardless of performance. A forgotten item does not deserve a longer gap simply because the timetable says so.
• Keeping gaps tiny because success feels good. Easy repetition can preserve fluency while providing weak evidence of durability.
• Ignoring the final retention interval. A schedule for Friday’s quiz is not automatically a schedule for a national examination three months later.
• Counting rereading as retrieval. Seeing the answer again changes the exposure history, not necessarily independent access.
• Scheduling the topic rather than the capability. Recognition, recall, explanation and transfer may need different returns.

14. What the Research Does Not Justify

Research does not justify a universal formula such as “1 day, 3 days, 7 days, 14 days” for every learner and every subject.

Nor does it justify the opposite claim that schedule shape never matters. Some experiments find differences under particular materials, interference patterns, populations and retention intervals. For example, expanding practice has shown advantages under high interference in text learning, while other work finds equivalent long-term retention with equal intervals. See Storm, Bjork and Storm, 2010.

The evidence points toward conditional optimisation rather than one sacred schedule.

15. Parent and Tutor Guide

When a learner uses spaced practice, ask four questions:

• What capability is being retrieved?
• How long must it survive?
• Was the last retrieval genuinely independent?
• What changed when the gap became longer?

If the learner repeatedly fails, do not respond with endless immediate repetitions alone. Diagnose whether the problem is weak initial understanding, poor cues, interference, missing feedback or a gap that expanded too aggressively.

If the learner succeeds too easily, change the cue, increase delay or require a more demanding use of the knowledge.

16. Delayed Independent Check

To test whether a spacing schedule built durable learning:

1. Wait beyond the learner’s normal review interval.
2. Use a fresh cue.
3. Remove notes and answer options where appropriate.
4. Require the knowledge inside a new problem or explanation.
5. Score correctness and independence separately.

If the learner succeeds only when the original flashcard cue appears, the schedule maintained a route, not necessarily portable knowledge.

17. Return: Schedule Shape Is a Control Variable, Not a Commandment

Spacing works because time changes the retrieval problem.

The next refinement is not to worship one geometric pattern of gaps. It is to use time deliberately.

Choose the retention horizon. Retrieve. Watch success. Widen when evidence permits. Repair when it does not. Change the retrieval job as knowledge develops. Then test whether the capability survives after the calendar stops helping.

Continue through How Spaced Practice Works, How Successive Relearning Works, Time-Variant Forgetting, the How Studying Works Numbered Series Reading Index and the How X Works Hub.