Module 1 · Replace rereading with testing and spacing
Lesson 1 of 6 · 6 min · English · Read-only
The fluency illusion is believing you know something because it reads smoothly and looks familiar, so judge what you know by what you can write or say with the page closed, not by how the page feels.
Rereading feels like learning because each pass is easier than the one before, and people tend to read ease as knowledge. But a test, a conversation or a job asks for something different: producing the content with the page closed. Familiarity and the ability to produce are separate things. Rereading reliably builds the first.
Take the running example of this course: a biology chapter on cell parts. With the page open, every line prompts the next. You see “the Golgi apparatus modifies, sorts and packages proteins” and think “yes, I know that.” What you did was recognize a sentence. The page did the remembering. Pulling something out of memory with nothing in front of you is called retrieval, and it is a different act. This course calls the false confidence that comes from recognizing instead of retrieving the fluency illusion: believing you know something because it reads smoothly and looks familiar.
The best-known test of this is a 2006 experiment by Roediger and Karpicke. Students studied short science passages, then either studied a passage again or wrote down everything they could recall from it, with no feedback. A final recall test came 5 minutes, 2 days or 1 week later. At 5 minutes, the group that had studied again did better. At 2 days and at 1 week, the group that had practised recalling did better. In a second experiment, the students who had reread most were the most confident they would remember the passage a week later, and of the three groups they remembered the least.
Read that as strong evidence about what it tested, not a promise about everything: the students were undergraduates at one university, the material was short passages, and the test was free recall.
A 2013 review of ten study techniques by Dunlosky and colleagues, which this course calls the Dunlosky paper, points the same way. It rated practice testing and spacing high in usefulness, and rereading and highlighting low. Its main reason for rating rereading low was the head-to-head comparison: against practice testing and similar methods, rereading consistently lost. Rereading does beat reading a text once in many studies, mostly on tests soon afterwards. Highlighting, as students usually do it, did little to boost performance in most situations the paper examined.
You read a one-page summary of cell parts three times and rate how well you know it: 8 out of 10. Then you run the closed-page check, but on five named parts only: the nucleus, the mitochondrion, the ribosome, the Golgi apparatus and the lysosome. You write what each one does. You get the nucleus and the mitochondrion right, and you have nothing for the ribosome, the Golgi apparatus or the lysosome. All of these numbers are assumptions of the example.
Predict before the next step
Which would you trust to decide what to study next, the 8 out of 10 or the written result? And would you spend the next session rereading the whole page, or doing something else?
Your writing stays on this page only. It is not saved or assessed.
Put both numbers on one scale. Read the rating as a claim: “I know about 80 percent of this page.” The check says 2 out of 5, which is 40 percent. They disagree by 40 percentage points, and the check is the one that looks like the exam: page closed, answer produced.
Look at which items were missed. The ribosome, the Golgi apparatus and the lysosome are the three the learner only recognized. With the page open each sentence looked familiar, so three readings felt like three passes over knowledge. A fourth full reread would spend part of its time on the two parts already known, and those are the lines that read most smoothly.
A better use of the same time: study only the three missed parts, close the page, and write them again. If all three come out, the learner has moved them from recognized to retrieved.
Five parts are a small sample, and a real check covers the whole page. The pattern is what matters here: the feeling ran ahead of the result.
Result
the written 2 out of 5 replaces the felt 8 out of 10, and the next session goes to the ribosome, the Golgi apparatus and the lysosome, not to the whole page.
Think it over
Pick a page you reread this week, close it and write what it said: how much came out, and what would you change about how you studied it?