WordieThe Testing Effect Explained: Why Trying to Remember Beats Studying Again
Tests are usually treated as a way to measure learning. A large body of research shows they are also one of the most powerful ways to cause it, and vocabulary is where the effect is easiest to use.
- The testing effect (also called retrieval practice) is the finding that recalling information from memory strengthens it more than restudying it for the same time.
- Roediger and Karpicke (2006) found that restudying won on a test five minutes later, but testing won clearly after two days and one week.
- Karpicke and Roediger (2008) showed that once Swahili words were learned, continuing to test them drove long-term retention, while continuing to restudy them added little.
- Dunlosky et al. (2013) rated practice testing as one of only two "high utility" study techniques, alongside distributed practice.
- For vocabulary, the most effective form is producing the meaning (or the word) from memory before checking, then spacing those attempts over days.
The testing effect is the finding that pulling information out of memory strengthens it far more than putting it in again. If you spend five minutes trying to recall the meanings of ten new words, you will remember more of them next week than if you spend the same five minutes rereading the definitions. Decades of experiments support this, and it is one of the cheapest improvements any learner can make: close the book, try to remember, then check.
What exactly is the testing effect?
In memory research, a "test" is any act of retrieval practice: answering a question, filling a gap, recalling a definition, or writing down everything you remember about a topic. The testing effect is the observation that this retrieval does more than measure memory. It changes it, making the retrieved information easier to access in the future.
The idea is old. Educational researchers documented benefits of recitation in the early twentieth century. But it was a series of studies at Washington University in St. Louis in the 2000s that brought it into mainstream educational advice, partly because they showed something counterintuitive: students consistently believed rereading worked better, even when the data showed the opposite.
The key studies
Roediger and Karpicke (2006): test-enhanced learning
Henry Roediger and Jeffrey Karpicke had college students read short prose passages on scientific topics. After an initial reading, students either read the passage again or took a free-recall test, writing down as much as they could remember without feedback. They were then tested after five minutes, two days or one week.
The result was a crossover. At five minutes, the students who restudied did slightly better. After two days and especially after one week, the students who had taken a test remembered clearly more. A second experiment compared repeated studying with repeated testing and found the same pattern, while students in the repeated-study condition predicted they would remember more. They were wrong.
This is the core lesson: rereading produces a short-lived feeling of fluency; retrieval produces durable memory. The difference only becomes visible after a delay.
Karpicke and Roediger (2008): the critical importance of retrieval
This study used something very close to vocabulary learning: 40 Swahili–English word pairs (for example mashua – boat). Students went through alternating study and test periods until they had learned all the pairs. The manipulation came after each pair was first recalled correctly:
- In some conditions, a learned pair was dropped from further study but kept in the tests.
- In other conditions, it was dropped from further testing but kept in the study periods.
All groups reached the same level during learning. One week later, the difference was dramatic. Students who kept being tested on learned pairs recalled around 80% of them. Students who stopped being tested once a pair was learned recalled only around a third, even if they kept studying it. Extra study of already-learned items added almost nothing; extra retrieval was what made the words stick. And once again, students' predictions of their own performance did not reflect this.
Dunlosky et al. (2013): ranking study techniques
John Dunlosky and colleagues reviewed the evidence for ten popular study techniques in Psychological Science in the Public Interest. They judged each on how well it generalized across learners, materials and settings. Only two earned a "high utility" rating: practice testing and distributed practice (spacing). Rereading, highlighting and summarizing were rated low utility. For a vocabulary learner, that verdict is direct: the two techniques that matter most are the testing effect and the spacing effect, ideally combined.
Later meta-analyses, such as Rowland (2014) in Psychological Bulletin, confirmed that the testing effect is robust, though its size varies. It tends to be larger when tests require recall rather than recognition and when feedback is given.
Why does retrieval strengthen memory?
Researchers still debate the mechanism, but several explanations have support:
- Retrieval effort. Harder (but successful) retrievals produce bigger gains. This fits Bjork's idea of desirable difficulties and explains why recalling after a delay helps more than recalling immediately.
- Elaboration and new routes. Searching memory activates related information, so the target becomes linked to more cues.
- Practice matching the use. In real life you need to retrieve words, not recognize them on a page you just read. Testing practices that exact operation.
- Better metacognition. A test shows you what you do not know, so you spend the next session on the right items.
Recall vs recognition: which kind of test?
Not all tests are equal. Multiple-choice and "do I recognize this?" checks are easier and still beneficial, but tasks that make you generate the answer usually produce more durable learning.
| Format | Example | Effort | Best use |
|---|---|---|---|
| Recognition | Pick the meaning of terse from four options | Low | Very new words, quick warm-ups |
| Receptive recall | See terse, say "brief and slightly unfriendly" | Medium | Core review for reading vocabulary |
| Productive recall | See "using few words, almost rudely", produce terse | High | Words you want to use in writing and speech |
| Cloze in context | "Her reply was ____: 'No.'" | High | Building usage and collocation knowledge |
A cloze deletion combines retrieval with context, which also addresses one of the main weaknesses of bare word-definition cards (see why flashcards alone fail). You can make cloze cards from your own sentences with the cloze deletion generator.
How to apply the testing effect to vocabulary
- Always attempt before you look. When reviewing a word, pause and genuinely try to recall the meaning. Say it aloud or type it. Only then reveal the answer.
- Get feedback immediately. Check the answer every time. Tests without feedback can let errors stick.
- Rate honestly. Distinguish "knew it instantly", "got it with effort" and "missed it". Easy items can wait longer; missed ones should come back soon.
- Keep testing words you know. Do not retire a word after one success. Space later tests out instead.
- Move from recognition to production. Once you reliably recall the meaning, start producing the word from its meaning or from a sentence gap.
- Use real sentences. Test yourself on the sentence where you found the word, not just the dictionary gloss.
Wordie's recall mode is built around this sequence: you see a saved word, try to recall its meaning, reveal the answer and rate it Forgot, Hard or Easy, which updates a mastery score for that word. It is a simple tool rather than a full scheduling algorithm, but it puts retrieval, not rereading, at the center of review.
Common objections
"Testing is stressful"
The benefit does not depend on stakes. Self-quizzing at home, with no grade and no audience, produces the effect. Some studies even suggest regular low-stakes practice testing can reduce anxiety before real exams, because the material feels more secure.
"I don't know the words well enough to be tested yet"
You need some initial encoding first; testing yourself on words you have never seen is just guessing. But the threshold is lower than most people think. One look at a word and its sentence is often enough to make a recall attempt worthwhile the next day. Research on "pretesting" and errorful generation suggests that even unsuccessful attempts, followed by feedback, can improve learning.
"Rereading feels like it works"
It does feel that way, which is exactly the illusion these studies keep documenting. Rereading makes text familiar, and familiarity is easily mistaken for knowing. The only fair judge is a delayed test.
The bottom line
The testing effect is one of the strongest, most practical findings in learning science. Retrieving a word's meaning from memory strengthens it far more than looking at it again, the advantage grows with time, and it is largest when retrieval is effortful, followed by feedback, and spaced over days. If you change only one habit, make it this: when you review vocabulary, try to recall before you look. For the timing side of the equation, read our guide to spacing effect research, and for the bigger picture see how to remember new words.
Sources: Roediger, H.L. & Karpicke, J.D. (2006), "Test-enhanced learning: Taking memory tests improves long-term retention", Psychological Science; Karpicke, J.D. & Roediger, H.L. (2008), "The critical importance of retrieval for learning", Science; Dunlosky, J., Rawson, K.A., Marsh, E.J., Nathan, M.J. & Willingham, D.T. (2013), "Improving students' learning with effective learning techniques", Psychological Science in the Public Interest; Rowland, C.A. (2014), "The effect of testing versus restudy on retention: A meta-analytic review of the testing effect", Psychological Bulletin; Bjork, R.A. (1994), "Memory and metamemory considerations in the training of human beings".
Frequently asked questions
What is the testing effect?
The testing effect is the finding that actively retrieving information from memory, for example by answering a quiz question or recalling a word's meaning, leads to better long-term retention than spending the same time rereading or restudying the material.
Is the testing effect the same as active recall?
They describe the same idea from different angles. Active recall and retrieval practice are the study techniques; the testing effect is the research finding that those techniques improve memory.
Do tests need to be graded to work?
No. Low-stakes self-tests such as flashcards, recalling a definition aloud or writing down everything you remember work well. Grading adds pressure but not memory benefit.
Does it matter if I get the answer wrong?
Failed attempts still help as long as you see the correct answer afterwards. Feedback turns an error into a learning opportunity, and later research suggests even unsuccessful retrieval attempts can improve subsequent learning.
How do I use the testing effect for vocabulary?
Cover the definition and try to recall it before looking, or look at a definition or sentence gap and try to produce the word. Rate how hard it was, and space the next attempt further out when recall was easy.
Build your vocabulary while you browse
Wordie saves any word in one click, highlights it on every page you read, and quizzes you with active recall. Free, private, no account.
Add Wordie to Chrome — it's free
