How to Study, According to the Evidence

In 2013, five cognitive psychologists led by John Dunlosky did something unusual: they reviewed the evidence behind ten of the most common study techniques and published a ranking. The results should have changed every study hall in the country. The two techniques students rely on most — highlighting and rereading — landed at the bottom. The winners were techniques most students rarely use deliberately.

Here’s what actually works, what doesn’t, and how to build a routine from it.

The two that work best

Practice testing (retrieval practice)

Every act of recalling information strengthens it more than re-exposing yourself to it. In Roediger and Karpicke’s classic experiments, students who read a passage once and then repeatedly tested themselves recalled far more a week later than students who reread the passage four times — even though the rereaders felt more confident. Karpicke and Blunt (2011) found retrieval practice even beat elaborate concept-mapping.

In practice: close the book and write down everything you remember. Turn headings into questions and answer them. Use flashcards. Do practice problems without the solution visible. The struggle is not a sign it’s failing — the struggle is the mechanism.

Distributed practice (spacing)

The same hours, spread out, produce far more durable learning than the same hours bunched together. This is the spacing effect, documented across hundreds of studies. Three one-hour sessions across a week reliably beat one three-hour session, and the advantage grows with time-to-test.

In practice: start earlier and study shorter. Review material the day after you learn it, then again days later. For anything fact-dense, let a spaced-repetition app schedule each item individually.

The strong supporting cast

• Interleaving. Mixing problem types within a session (ABCABC) beats blocking them (AAABBB), especially in math and science — Rohrer and Taylor (2007) found large advantages on delayed tests. Blocked practice feels smoother; interleaved practice teaches you to recognize which approach a problem needs, which is the actual exam skill.
• Elaborative interrogation. Ask “why is this true?” and answer it. Facts connected to reasons behave like structures, not trivia.
• Self-explanation. After a worked example, explain each step to yourself. Where the explanation stumbles is exactly where your understanding has a hole.

What the evidence demoted

• Highlighting. Essentially no benefit in most studies — it marks text without processing it, and can even hurt by fragmenting reading.
• Rereading. Produces fluency, and fluency masquerades as knowledge — the illusion of competence. The second read feels easier, and you credit yourself with learning. The delayed test disagrees.
• Summarizing helps only when done well, and most students were never taught how.

None of these are sins. But every minute spent on them is a minute not spent on testing yourself, which the evidence says is worth several of those minutes.

The infrastructure: sleep and attention

Two boring factors outweigh any technique:

• Sleep is when memory consolidates. Newly encoded memories are stabilized and reorganized during sleep, and sleep deprivation measurably impairs both learning before it and retention after it. An all-nighter is a memory-destruction ritual with extra steps: you’re trading the consolidation step for a few more hours of low-quality encoding.
• Attention is the gate. Divided attention during encoding cripples memory formation. Thirty minutes with the phone in another room beats ninety minutes of half-studying — not as a moral claim, as a measured one.

A routine that follows the evidence

1. First contact: learn actively — take sparse notes in your own words, ask why as you go.
2. Same day: convert what must be remembered into questions — flashcards, problem sets, blank-page recall. This is the step that replaces rereading.
3. Following days: short, spaced review sessions. Answer the questions; don’t revisit the source unless you miss.
4. Mix it up: interleave topics within sessions rather than dedicating whole days to one subject.
5. Protect sleep the whole way through — especially the nights right after learning and right before the test.

The honest summary of a century of research fits in one sentence: test yourself, space it out, and sleep. Everything else is refinement. The spacing part is the piece humans reliably fail to do by hand — deciding which card, which day — and it’s the part Quizatto’s scheduler does for you, so your effort goes where the evidence says it counts.