How quizzes help people learn
You've just given a 45-minute talk. You run three questions to wrap up. Half the audience gets wrong something you spent 20 minutes explaining. The intuition is often that they weren't paying attention. A more likely explanation: they thought they had understood it, and the impression didn't survive the week it took to test it.
Cognitive psychology has replicated this effect many times. Retrieving information strengthens memory while checking it, and can be especially useful when participants feel familiar with the material already.
The testing effect: a well-documented finding rarely believed in practice
In 2006 Henry Roediger and Jeffrey Karpicke published a study that became a foundational reference for retrieval-practice research. Students read a text. Half of them read it four times. The other half read it once and took three retrieval tests, trying to remember the content without looking at the text.
Right after, the rereading group did better. The students themselves thought they had learned more from rereading. A week later the picture was reversed: the test group remembered substantially more, even though the rereading group had been more confident in their own retention.
Karpicke and Janelle Blunt extended the finding in a 2011 Science study: they compared retrieval practice with concept mapping, a widely used teaching technique. Retrieval practice won. On the final test a week later, the retrieval group was about 50 percent higher, and the advantage held for both verbatim questions and inference questions that required understanding.
Why the feeling misleads
The surprising result is the gap between confidence and retention. In Roediger and Karpicke's 2006 study, the rereading group was most confident they'd remember the material, even though the later test showed that the retrieval-practice group had retained more.
Robert Bjork, Jeffrey Dunlosky and Nate Kornell explain the mechanism in Annual Review of Psychology (2013). The brain uses fluency as an indicator of understanding. When you read a passage a second time, the sentences are familiar, and familiarity gets mistaken for knowing the material. But familiarity tells you something about the text, not about what you can produce without the text in front of you.
The error has a property that makes it especially persistent: if rereading gives you a sense of understanding, you don't have much reason to check further. What would have revealed the understanding to be shallow, trying to produce the information without the text, is exactly the effort that rereading removes the need for.
Bjork calls this class of tasks "desirable difficulties": tasks that feel harder in the moment but give better retention. The effort involved appears to be tied to the mechanism itself. This is also why the structure set by the presenter or instructor matters. Left to choose, many participants reread because it feels easier in the moment.
Why wrong answers are worth the time
A common objection to testing is: "what if people get it wrong?" Kornell, Hays and Bjork (2009) found that attempts to retrieve information participants had not yet learned actually improved learning, provided the correct answer was presented afterward.
Struggling to come up with an answer, even without success, seems to make the brain more receptive to the correct answer when it arrives. This is the foundation of pretesting: ask the question before you give the answer. An unsuccessful retrieval attempt can support later learning when the correct answer follows.
What works in practice
1. Pretest before you explain
Ask a question about something you haven't explained yet. Most people will get it wrong, and that's the point. A wrong answer followed by the right one sticks better than the right answer alone. It takes one minute and improves both engagement and retention.
2. Add micro-quizzes during the session
Short quizzes during the session tend to help retention more than a single closing test. Szpunar and colleagues (2013) found that interpolated tests during video lectures reduced mind wandering and improved retention. You break passive listening and give the brain a chance to consolidate along the way. It also prevents the sense of having understood from building up across a whole session without being tested.
3. Anonymous answers lower the threshold
Test anxiety is about consequences. In Agarwal et al. (2014) 72 percent of students reported that frequent, low-stakes quizzes made them less nervous about tests. This is self-report, not a measured anxiety score, but the direction is worth noting: normalizing testing shifts the framing away from judgment. Anonymity lowers the threshold further.
4. Spend a little time on wrong-answer analysis
When you show the distribution of answers and many got it wrong, participants realize they aren't alone. It also gives you a concrete starting point for explaining the specific misconception. Moving on immediately after the right answer is a common mistake; much of the learning sits in spending a minute on why the other option was tempting.
The limits of the finding
Retrieval practice isn't a universal solution. The strongest effects are found for factual knowledge, definitions, and concepts with clear answers. For open skills such as critical thinking, writing technique, and creative problem solving, the picture is more mixed in our reading. Other techniques (spacing and worked examples) appear to matter at least as much.
Multiple-choice testing in particular has a less advertised downside documented by Roediger and Marsh (2005): seeing a plausible wrong answer can, under some conditions, lodge that wrong answer in memory. Most of the testing-effect research is built on short-answer and free-recall formats, where you produce the answer rather than recognize it from a list. If you lean heavily on multiple choice, you get some of the benefits of retrieval practice and some of the risks of false-alternative exposure. Roediger and Marsh gave no feedback in their study. Butler and Roediger (2008) found that feedback reduces the effect, so show the right answer after each question.
It's also worth distinguishing consolidation from discovery. Retrieval practice works on material the learner has already encountered; it's how that material gets secured. It isn't a substitute for the initial work of making sense of something new. A session that is all quizzing, with no exposition, teaches nothing.
What holds across these caveats is a simple principle: what we can retrieve from memory sticks better than what we've just recognized on a page. Recognition is a common but relatively unreliable signal of what you actually know.
We often picture learning as information going into the brain: we read, listen, or watch a video, and expect the knowledge to settle into place. For much of what we need to remember, retrieving that information again strengthens the memory, often through effort.
Rereading can create a feeling of mastery that discourages further checking. Placing quizzes throughout a session makes retrieval part of the work with the content while there is still time to address gaps.
Common questions
Why does it feel like we learn more by rereading?
Because fluency is pleasant. When you read a passage a second time, you recognize the sentences, and recognition is easily confused with understanding. Bjork describes how fluency creates illusions of learning. Retrieval practice feels harder and gives better retention a week later; that effort contributes to the effect.
Why doesn't rereading give the same effect?
Rereading gives the brain the signal "this is familiar," which is easily read as "I know this." Familiarity comes from seeing the text, while retrieval asks what you can produce without it. The effort exposes gaps in memory while there is still time to address them.
Does retrieval practice also help with skills?
Yes, with nuance. Retrieval practice is best documented for factual knowledge and concepts. Distributed practice tends to matter more for motor skills. For complex skills such as analysis and argumentation, retrieving relevant concepts before using them can be a useful warm-up alongside other forms of practice.
Can quizzes make participants anxious?
High-stakes tests, such as final exams that determine grades, can cause anxiety. In Agarwal et al. (2014), however, 72 percent of students reported that frequent, low-stakes quizzes made them less nervous about tests. This is self-report rather than a controlled anxiety measure, but it suggests that routine low-stakes practice can shift the experience away from judgment. Anonymous quizzes can lower the threshold further.
Sources
- Roediger, H. L. & Karpicke, J. D. (2006). Test-enhanced learning: Taking memory tests improves long-term retention. Psychological Science, 17(3), 249–255.
- Karpicke, J. D. & Blunt, J. R. (2011). Retrieval Practice Produces More Learning than Elaborative Studying with Concept Mapping. Science, 331(6018), 772–775.
- Bjork, R. A., Dunlosky, J. & Kornell, N. (2013). Self-regulated learning: Beliefs, techniques, and illusions. Annual Review of Psychology, 64, 417–444.
- Kornell, N., Hays, M. J. & Bjork, R. A. (2009). Unsuccessful retrieval attempts enhance subsequent learning. Journal of Experimental Psychology: Learning, Memory, and Cognition.
- Agarwal, P. K., D’Antonio, L., Roediger, H. L., McDermott, K. B. & McDaniel, M. A. (2014). Classroom-Based Programs of Retrieval Practice Reduce Middle School and High School Students’ Test Anxiety. Journal of Applied Research in Memory and Cognition, 3(3), 131–139. (The measured intervention was clicker-based classroom quizzes.)
- Szpunar, K. K., Khan, N. Y. & Schacter, D. L. (2013). Interpolated memory tests reduce mind wandering and improve learning of online lectures. PNAS, 110(16), 6313–6317.
- Roediger, H. L. & 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.
- Butler, A. C. & Roediger, H. L. (2008). Feedback enhances the positive effects and reduces the negative effects of multiple-choice testing. Memory & Cognition, 36(3), 604–616.
Make quizzing part of the session
Plenora lets you build anonymous quizzes directly into slides, show the answer distribution live, and reuse questions across sessions. Use them to strengthen retrieval while participants work through the material.