Retrieval practice (active recall)
How learning works · Published · Evidence: Strong · About this guide
Retrieval practice, often called active recall, means getting what you studied back out of memory, by answering questions or writing down what you remember, instead of reading it again. The benefit is known as the testing effect: in two large meta-analyses practice tests beat restudying by about half a standard deviation, and in classrooms quizzing raised achievement by a similar amount [1][2][3]. These are group averages, not a forecast for each student.
Most studying puts information in: reading notes again, highlighting, watching a lecture a second time. Retrieval practice works the other way round. You close the book and try to get the information back out, by answering questions, writing down what you remember, explaining an idea from memory or going through flashcards. Researchers call the benefit that follows the testing effect, although the "test" can be anything that makes you recall, with no grade attached. This article covers what studies have found, how large the effects are, and where the evidence runs out.
Is retrieval practice the same as active recall?
Yes. Active recall is the name many study guides use for the same habit: getting information back out of memory instead of putting it in again. Retrieval practice is the researchers' name for the habit, and the testing effect their name for the benefit that follows.
What did the classic experiment show?
In a 2006 study, university students learned short factual passages, one about the sun and one about sea otters [4]. Some spent four 5-minute sessions reading a passage. Others read it once and then spent three 10-minute sessions writing down everything they could remember, without being told what they had missed. Half of the students took a final recall test five minutes later, the other half a week later.
Five minutes after studying, the readers did best: they recalled 83% of the passage's ideas, against 71% for the students who had practised recalling. A week later the order had reversed: students who had practised recalling produced 61% of the ideas, students who had only read produced 40%. A first experiment, which gave reading and recalling the same time, showed the same reversal after two days and after a week.
The reading group was also the more confident of the two that it would remember the passage. A method that feels better and works worse over time is a pattern this guide returns to in the article on illusions of knowing.
Rereading or recalling: ideas remembered after 5 minutes and after a week (Roediger & Karpicke 2006)
Measured data from Roediger and Karpicke (2006), Experiment 2. Five minutes after studying, the students who had spent four study periods reading the passage recalled the most of its ideas (83%); a week later they recalled the least (40%), while those who had read it once and recalled it three times recalled 61%. The readers were also the surest, on a 1–7 scale, that they would remember it. Each recall bar is a separate group of 30 students; group averages from one study, not a prediction for any one learner.
Data: Roediger & Karpicke (2006), Psychological Science 17(3), Experiment 2, p. 252 and Table 2. [4]
Data table
| What was measured | Four study periods | Three study periods, one recall test | One study period, three recall tests |
|---|---|---|---|
| Share of the passage's ideas recalled: After 5 minutes | 83% | 78% | 71% |
| Share of the passage's ideas recalled: After a week | 40% | 56% | 61% |
| Expected recall a week later (1–7): Rating after studying | 4.8 | 4.2 | 4.0 |
- Study periods lasted 5 minutes and recall tests 10 minutes. Experiment 1, which gave reading and recalling the same time, showed the same reversal.
- After a week, the authors call the difference between one and three recall tests (56% and 61%) marginal.
- The rating asked how well the students thought they would remember the passage in a week (1 = not very well, 7 = very well), at the end of the study session.
How large is the effect across many studies?
A single experiment can mislead, so researchers pool many. Three large meta-analyses point the same way:
- A 2014 analysis of 159 results from 61 studies compared practice tests with restudying. The average advantage was about half a standard deviation (g = 0.50) [1].
- A 2017 analysis of 272 results found practice tests ahead of restudying (g = 0.51) and further ahead of doing nothing or an unrelated task (g = 0.93) [2].
- A 2021 analysis limited to real classrooms, with data from about 48,000 students in 222 studies, found that quizzing raised achievement by a medium amount on average (g = 0.50) [3].
These figures describe a shift in a group's average, not what happens to each student. A 2013 review of ten common study techniques rated practice testing as one of only two with high usefulness; the other was spreading study out over time [5].
Does it only work for memorising facts?
The evidence says no, though it is thinner here. In a 2011 experiment, students read a science text once and then spent the same total time either on two rounds of recall with a rereading in between, or on building concept maps, diagrams that link the ideas, while looking at the text [6]. A week later, the recall group did better both on questions answered directly in the text and on questions that required drawing inferences. In a second experiment, 101 of 120 students did better on the text they had practised recalling, and the advantage held when the final test asked them to draw a concept map. A published comment questioned the study's methods and its assumptions about learning, and the authors replied [7][8]. A preregistered 2023 study with 230 participants found the same advantage on questions answered directly in the text, but it disappeared when a concept-mapping group, given the same total time, spent its last five minutes memorising its map. The original advantage may therefore owe more to the extra rereading built into the recall condition than to recall itself. On that study's two inference questions, the concept-mapping groups together scored higher than the recall group [9]. In classrooms, quizzing helped with understanding concepts and applying them to problems, as well as with facts [3].
How far the benefit reaches beyond the questions you practised is a separate matter. A 2018 meta-analysis found that practice testing can help with different but related questions, especially ones asking for application or inference. Transfer was among the weakest to material that was studied but not tested, and after adjusting for results that may have gone unpublished, the estimated transfer often fell to none when none of three helpful conditions was present: practice and final questions calling for the same answer, practice that went beyond recalling the bare answer, and high success during practice [10].
Does it matter whether you get the answers right?
It does. Without feedback, practice mostly strengthens what you manage to recall. In the 2014 meta-analysis, studies without feedback in which people answered half or fewer of the practice questions correctly showed no reliable benefit. Studies that showed the material again after the test had larger effects (g = 0.73, against 0.39 without feedback) [1]. The classroom analysis also found larger gains when quizzes came with corrective feedback [3], although the 2017 analysis found no difference [2].
Wrong answers can also stick. After a multiple-choice test on facts from short passages, students later gave some of the wrong options they had read as answers, even though they had been warned not to guess, and the more wrong options they had read, the more often this happened. The test still improved their recall overall [11]. Feedback after the test raised later correct answers and cut those intrusions [12]. A mistake becomes useful for learning when you then find out the right answer; the article on feedback and mistakes covers this.
Are multiple-choice questions good enough?
A common criticism is that multiple choice asks you to recognise the answer rather than recall it. In laboratory studies, recalling an answer from a cue gave larger benefits than recognising it. Writing down everything you could remember clearly beat recognition where the answers were checked afterwards or were mostly right during practice [1]. In classrooms, quizzes that asked students to recognise the answer, almost all of them multiple choice, helped as much as quizzes that asked them to recall it [3].
The wrong options may matter. In two experiments, students answered questions whose wrong options were plausible, for example four real geysers, only one of them the tallest [13]. Multiple-choice practice helped later recall of the tested facts as well as answering the same questions without options did. It also did better than practice without options on related questions whose answers had appeared as wrong options; the gain over unpractised questions was reliable only in the larger second experiment. These were two small laboratory studies with the final test a few minutes after practice, so they say little about how long that extra benefit lasts.
Why does it feel less effective than rereading?
Rereading makes material feel familiar, and familiarity is easy to mistake for knowing. In the 2006 and 2011 experiments described above, students predicted the opposite of what happened: they expected rereading or concept mapping to serve them better than recall [4][6]. Struggling to remember can feel like failing even while it is helping.
How can you use it?
The studies suggest habits, not a formula. Each of these is a way to practise active recall:
- Flashcards. A cue on one side, the answer on the other: try to recall the answer before you turn the card. In laboratory studies, recalling an answer from a cue helped more than recognising it among options [1].
- Blank-page recall ("blurting" in some UK study guides). After reading or watching something, close it and write or say what you remember before looking again, as the students in the classic experiment did in writing [4].
- Practice questions and old exams. Answer them before looking at the answers, then check every one: tests helped more in studies that showed the material again afterwards [1].
- Come back to the same material more than once. In classrooms, quizzing the same content more often went with larger gains [3], although the 2017 analysis, mostly of laboratory studies, found larger effects from one practice test than from several [2]. When to come back is the subject of the article on spaced practice.
- A quiz does not have to count toward a grade to help: in the classroom analysis, low-stakes quizzes helped about as much as high-stakes ones [3].
Whether practice tests also make exams feel less stressful is a different question, taken up in the article on exam stress and memory.
What the research does not say
- It is not a guarantee for every person. The figures above are averages over studies whose results varied widely; your own result may differ.
- It does not beat every active method. In classrooms, quizzing clearly beat rereading (g = 0.33) and doing nothing extra, but against other active strategies such as concept mapping, note-taking or summarising, the difference was small and not statistically reliable (g = 0.10), in relatively few studies [3]. In the laboratory, an advantage over concept mapping disappeared once the time spent memorising was matched [9].
- Not every test helps. Low success with no feedback showed no reliable gain [1], and wrong options that nobody corrects can be learned as facts [11].
- Practising some questions does not cover a whole subject. The benefit is largest for what you practised; the spill-over to untested material is smaller and less certain [10][3].
- How you do right after studying is a poor guide to later. In the 2006 study, the students who had only reread scored best after five minutes and worst a week later [4].
- Most of the evidence comes from students. Most laboratory studies used university students [1]. In classrooms the effect appeared in elementary, middle and high school and at university, with no reliable difference between these levels; there were too few studies of adult continuing education to draw a conclusion [3]. Published laboratory studies also reported larger effects than unpublished ones [1], while the classroom analysis judged the risk of such bias small [3].
- It does not give you a schedule. How often and when to practise are the subjects of the articles on how much practice is enough and on spaced practice.
Related articles
-
Feedback and learning from mistakes
Why checking answers after a quiz matters, when a wrong guess helps you learn, and why confident mistakes are often easiest to fix, though they can return.
-
Spaced repetition and spaced practice
The spacing effect: why spreading study over days helps memory last, how far apart reviews should be, and whether ever-longer intervals beat equal gaps.
-
How many times should you review something to remember it?
Getting it right once is a start, not the finish. What research says about practising to a criterion, successive relearning, overlearning and when to stop.
-
The illusion of competence: when studying feels better than it works
Familiar is not the same as remembered. Why rereading and highlighting feel effective but rate low in research, and how to check what you really know.
-
Elaborative interrogation and self-explanation: explaining things to yourself
Asking why a fact is true, or explaining each step, tends to help. What studies found, what it costs in time, and how it compares with testing yourself.
-
The pretesting effect: guessing before you learn
Answering questions before a lesson can help you remember those answers, if the correct answer follows. What pretesting studies found, and their limits.
Sources
- [1] Rowland, C. A. (2014). The effect of testing versus restudy on retention: A meta-analytic review of the testing effect. Psychological Bulletin, 140(6), 1432–1463. meta-analysis · doi.org/10.1037/a0037559 ↩ Back to the text
- [2] Adesope, O. O., Trevisan, D. A., & Sundararajan, N. (2017). Rethinking the use of tests: A meta-analysis of practice testing. Review of Educational Research, 87(3), 659–701. meta-analysis · doi.org/10.3102/0034654316689306 ↩ Back to the text
- [3] Yang, C., Luo, L., Vadillo, M. A., Yu, R., & Shanks, D. R. (2021). Testing (quizzing) boosts classroom learning: A systematic and meta-analytic review. Psychological Bulletin, 147(4), 399–435. meta-analysis · doi.org/10.1037/bul0000309 ↩ Back to the text
- [4] Roediger, H. L., III, & Karpicke, J. D. (2006). Test-enhanced learning: Taking memory tests improves long-term retention. Psychological Science, 17(3), 249–255. experiment · doi.org/10.1111/j.1467-9280.2006.01693.x ↩ Back to the text
- [5] Dunlosky, J., Rawson, K. A., Marsh, E. J., Nathan, M. J., & Willingham, D. T. (2013). Improving students' learning with effective learning techniques: Promising directions from cognitive and educational psychology. Psychological Science in the Public Interest, 14(1), 4–58. review · doi.org/10.1177/1529100612453266 ↩ Back to the text
- [6] Karpicke, J. D., & Blunt, J. R. (2011). Retrieval practice produces more learning than elaborative studying with concept mapping. Science, 331(6018), 772–775. experiment · doi.org/10.1126/science.1199327 ↩ Back to the text
- [7] Mintzes, J. J., Canas, A., Coffey, J., Gorman, J., Gurley, L., Hoffman, R., McGuire, S. Y., Miller, N., Moon, B., Trifone, J., & Wandersee, J. H. (2011). Comment on "Retrieval practice produces more learning than elaborative studying with concept mapping". Science, 334(6055), 453. comment · doi.org/10.1126/science.1203698 ↩ Back to the text
- [8] Karpicke, J. D., & Blunt, J. R. (2011). Response to comment on "Retrieval practice produces more learning than elaborative studying with concept mapping". Science, 334(6055), 453. comment · doi.org/10.1126/science.1204035 ↩ Back to the text
- [9] Mayrhofer, R., Kuhbandner, C., & Frischholz, K. (2023). Re-examining the testing effect as a learning strategy: the advantage of retrieval practice over concept mapping as a methodological artifact. Frontiers in Psychology, 14, Article 1258359. experiment · doi.org/10.3389/fpsyg.2023.1258359 ↩ Back to the text
- [10] Pan, S. C., & Rickard, T. C. (2018). Transfer of test-enhanced learning: Meta-analytic review and synthesis. Psychological Bulletin, 144(7), 710–756. meta-analysis · doi.org/10.1037/bul0000151 ↩ Back to the text
- [11] Roediger, H. L., III, & Marsh, E. J. (2005). The positive and negative consequences of multiple-choice testing. Journal of Experimental Psychology: Learning, Memory, and Cognition, 31(5), 1155–1159. experiment · doi.org/10.1037/0278-7393.31.5.1155 ↩ Back to the text
- [12] Butler, A. C., & Roediger, H. L., III (2008). Feedback enhances the positive effects and reduces the negative effects of multiple-choice testing. Memory & Cognition, 36(3), 604–616. experiment · doi.org/10.3758/MC.36.3.604 ↩ Back to the text
- [13] Little, J. L., Bjork, E. L., Bjork, R. A., & Angello, G. (2012). Multiple-choice tests exonerated, at least of some charges: Fostering test-induced learning and avoiding test-induced forgetting. Psychological Science, 23(11), 1337–1344. experiment · doi.org/10.1177/0956797612443370 ↩ Back to the text