Learning Strategies That Work: What the Research Shows

The best-supported strategies are retrieval practice (testing yourself) and spacing your study over several days instead of cramming at the last minute. Interleaving, elaboration and self-explanation look promising; highlighting and rereading, the most popular methods, fare poorly in the research. So “learning how to learn” does not start with new tricks. It starts with swapping comfortable habits for more effective ones.

How can you tell a learning strategy that works?

“Works” means it measurably improves what you still know and can apply later, not how confident you feel while studying. The difference is large, because methods that feel smooth often only create a sense of familiarity.

The key overview comes from Dunlosky, Rawson, Marsh, Nathan and Willingham, published in 2013 in “Psychological Science in the Public Interest”. The team rated ten common learning techniques by how broadly they work: across age groups, materials and kinds of tests. The result comes in three tiers:

• High utility: practice testing and distributed practice.
• Moderate utility: elaborative interrogation, self-explanation and interleaved practice.
• Low utility: summarization, highlighting and underlining, the keyword mnemonic, imagery use for text learning, and rereading.

“Low utility” does not mean “useless”. It means the evidence is not strong enough to recommend the technique widely, or that better alternatives exist for the same time. And “moderate” is meant honestly too: elaborative interrogation and self-explanation had not yet been adequately evaluated in real classrooms, and systematic research on interleaving had only just begun.

What is retrieval practice, and why does it rank first?

Retrieval practice means pulling knowledge out of memory instead of looking at it again: flashcards, quiz questions, a blank page on which you write everything you know. According to Dunlosky and colleagues, the effect holds across many test formats, kinds of material, learner ages and retention intervals. It costs little time compared with other techniques and needs hardly any training. In practice: close the book and ask yourself what you just learned.

Retrieval is not just a check; it is itself the learning event. For more and a how-to, see Active recall: retrieve instead of rereading.

How should you space your learning?

The same study time, spread over several sessions, beats one block right before the exam. This is the spacing effect, which Hermann Ebbinghaus already described in 1885 in “Über das Gedächtnis”. According to Wikipedia, the meta-analysis by Cepeda and colleagues (2006) found that spaced practice beat massed practice in 259 of 271 comparisons. Dunlosky and colleagues rate distributed practice as high utility because it works across age groups, materials and long delays.

In practice: instead of ten hours in one day, five sessions of two hours, and retrieve the material again after a few days. There are fixed systems for the rhythm, for example in the article on spaced repetition and in the Leitner system.

What does interleaving add?

Interleaving means practising different problem types or topics in alternation instead of working through one type in a row. Dunlosky and colleagues rate it “moderate”: it showed marked effects on learning and retaining mathematical skills, but the literature was still small and contained some null results. It feels harder because for each task you first have to recognise which method is required; that may be exactly where the benefit comes from. The article on interleaving shows how to use it.

How do you elaborate on material?

Elaboration means explaining new content and connecting it to what you already know. The review names two forms: elaborative interrogation (“Why is this so?”) and self-explanation (explaining how a solution step relates to the whole). Both received “moderate”. Their strength is that they worked across many materials and a wide age range; what remained open was mainly whether the effects hold for comprehension, over longer delays and in real classrooms; for elaborative interrogation, also whether it helps beyond simple fact lists with longer, more complex material.

In everyday study, ask “why” of every sentence in your notes and test whether you can explain a term in your own words. How exactly that works is covered in elaborative learning; a particularly rigorous variant is the Feynman technique.

And what about dual coding?

Dual coding means processing information verbally and visually at once, for example placing a diagram next to the explanatory text. Caution is in order here: in the overview by Dunlosky and colleagues it does not appear as a technique of its own. The technique they did rate, “imagery use for text learning” (picturing images while reading), received only “low utility”: it helps some learners on some tasks, but the conditions are narrow. If you use sketches, diagrams and mind maps, treat them as an aid to understanding and retrieval, not a miracle cure. For an introduction, see dual coding for studying; mind mapping is related as a form of visual organisation.

Which popular methods are overrated?

According to the overview, rereading and highlighting are among the methods students report using most, and both received “low utility”.

• Rereading: It is relatively cheap in time and needs little training, but, in the authors’ words, usually much less effective than other techniques; that disadvantage weighed most in the low rating. Instead of rereading, they recommend practice testing.
• Highlighting and underlining: In most situations studied it does little. It may help when you already know enough to highlight the right passages, or with difficult texts, and it can even hurt on tasks that require drawing inferences.
• Summarizing: It works well for learners who are already skilled at it; many need extensive training first.

That does not mean you may never highlight. Highlight sparingly as a signpost on the first pass, then turn the marked passages into questions. The same caution applies to learning styles: the idea that everyone learns best “in their type” is contested; more in the glossary under learning styles.

How do you build a routine from this?

You do not have to do everything at once. A lean routine for one exam topic:

1. Read once, thoroughly, and note questions instead of highlighting.
2. Retrieve from memory the same day: write down everything you know, then compare with the material.
3. Elaborate on the gaps: why is this true, how does it connect to other things?
4. Repeat at intervals: after a day, a few days, a week, each time retrieving again.
5. Mix topics: in later sessions, shuffle problems from different chapters.

Be honest with yourself: if studying feels effortful, that is usually a good sign for retrieval and spacing. Dunlosky and colleagues themselves note that students prefer the weaker methods, while in one study they cite, the more often students reported using practice testing, the better they did on the final exam.

Where can a tool help?

Retrieval and spaced repetition are exactly the parts that flashcards organise well. With the AI flashcard generator, LearnCastAI can create flashcards from your own materials, which you then retrieve; whether you actually master the material is shown only by the retrieval itself. The strategies also work with no tool at all, just paper and a calendar.

You will find more methods in the learning methods category.

Conclusion

If you want to change only two things: retrieve material actively instead of rereading it, and spread your study over several days. Everything else, interleaving, elaboration, sketches, is a sensible addition, not a replacement. And keep measuring yourself: if you test yourself regularly, you will quickly see what really sticks.

Start small: pick a single topic for this week, put two retrieval sessions in your calendar and watch how much you know the second time without help. That feedback is more telling than any gut feeling, and it shows which strategy makes the biggest difference for your material and your exam.