The spacing effect

Spacing works. When the same material is revisited across separated study occasions rather than clustered into one period, later retention is generally better. This result has been replicated across a large evidence base containing verbal materials such as paired associates, facts, paragraphs, spelling, names and texts.¹ Classroom and second-language meta-analyses also find benefits from distributed practice.²³

The figure shows the fixed-study-time comparison from one two-session experiment: adults learned obscure facts, and a fitted beneficial gap produced higher delayed recall than a zero-day gap.⁴

<meta http-equiv="Content-Security-Policy" content="default-src 'none'; style-src 'unsafe-inline'">
<style>
  .figure { font-family: system-ui, sans-serif; color: #111; border: 2px solid currentColor; padding: 1rem; }
  .title { font-weight: 700; margin-bottom: 1rem; }
  .row { display: flex; align-items: center; gap: .35rem; margin: 1rem 0; min-height: 3rem; }
  .label { width: 5rem; font-weight: 700; }
  .study { border: 2px solid currentColor; padding: .55rem; white-space: nowrap; }
  .later { color: #555; padding: .4rem; white-space: nowrap; }
  .stretch { flex: 1; border-top: 2px dotted currentColor; }
  .check { border: 3px double currentColor; padding: .55rem; font-weight: 700; animation: pulse 2.4s ease-in-out infinite alternate; }
  @keyframes pulse { from { opacity: .65; } to { opacity: 1; } }
</style>
<div class="figure" role="img" aria-label="Two-session obscure-facts experiment with fixed study time, comparing a zero-day gap with a fitted beneficial gap">
  <div class="title">Two-session obscure-facts experiment with fixed study time</div>
  <div class="row">
    <div class="label">Zero-day gap</div>
    <div class="study">Session 1</div><div class="later">same day</div><div class="study">Session 2</div>
    <div class="stretch"></div>
    <div class="check">Lower delayed recall</div>
  </div>
  <div class="row">
    <div class="label">Spaced gap</div>
    <div class="study">Session 1</div><div class="later">fitted beneficial gap</div><div class="study">Session 2</div>
    <div class="stretch"></div>
    <div class="check">Higher delayed recall</div>
  </div>
</div>

Figure: In one two-session obscure-facts experiment, a fitted beneficial gap improved delayed recall over a zero-day gap. Spacing is not the instruction to wait as long as possible. In a large two-session experiment, adults learned obscure facts using different gaps before final tests at different delays. Recall first improved as the study gap grew and then declined when the gap became longer.⁴ A gap can therefore be too short to provide useful separation or too long for that material and target test.

With a deck and twenty minutes, avoid spending the whole session cycling through the same small group of cards back-to-back. Give answered cards time away and return to them on later occasions. This applies the established spacing principle without pretending that one exact schedule has been validated for everyone.

Boundary. Spacing is strongly supported, but a universal interval or spacing ratio is not. Outcomes vary with the task, the interval and the delay before the final test.¹⁴ The older evidence base is heavily weighted toward young adults and tests conducted within a day, with long retention periods represented much less often.¹ Spacing justifies separation, not a single supposedly correct calendar.