Reading Speed and Comprehension

Comprehension is more than recognizing the words

You can recognize every word in a sentence and still misunderstand its point. Consider an argument that compares two proposals: understanding it includes knowing which proposal the author supports, why, and under what conditions the recommendation could change. Remembering a few repeated phrases is a different achievement.

Comprehension models distinguish building a representation from the words and integrating it with knowledge. The result should be coherent enough to support questions, explanations or further reasoning. [1]

That is why a reading goal should include an outcome. “Read the chapter” gives you a finish line but little guidance about what you need to retain. “Explain why the process changes when the temperature falls” gives you a question. You can then decide which definitions, steps and qualifications deserve attention.

Our recommendation is to replace a vague feeling of having understood with a small demonstration: explain the relationship, identify the evidence or apply the idea to a new case. This can reveal a gap while it is still easy to repair.

Four useful levels of understanding

Different tasks ask different things of the same text. A quick outline may be entirely adequate for deciding whether to read an article. It may be inadequate for using the article as the basis for a decision.

These categories are a practical way to plan your reading, not separate boxes in the brain. A good explanation may use several of them. The important point is that one easy recognition question cannot stand in for every kind of understanding.

What happens when you push the pace?

Less time leaves fewer opportunities to resolve an unfamiliar word, revisit a clause or connect a new idea with an earlier one. A reader may preserve the main theme while losing qualifications. In another passage the main theme itself may become unclear. The result depends on both reader and material.

Experiments comparing skimming and rapid sequential presentation illustrate why seeing more words does not necessarily mean building the same conceptual understanding. Presentation timing and the outcome tested are part of the comparison. [2]

There is no single words-per-minute threshold at which everyone’s comprehension switches off. A speed reached on familiar narrative text should not become a target for an unfamiliar technical explanation. Equally, a slow pace does not guarantee attention. Judge a pace by the quality of what you can do with the material.

When you want to experiment, change one thing at a time. Try a slightly brisker pace on a comparable new passage, then check the same kinds of questions. If the explanation becomes thin, the faster rate has not served that learning goal, even if the timer looks impressive.

A worked example: gist is not the whole message

A gist summary—“The library is staying open longer”—is correct as far as it goes. It misses who gains, who gives something up, how long the change is intended to last and the condition for continuing it. Whether that is an acceptable summary depends on your task.

• Finding a place to study: Which library and which days are affected?
• Explaining the decision: What funds the longer hours?
• Evaluating the policy: What would count as a successful trial, and whose access might be reduced?
• Avoiding an incorrect inference: Does the passage say all library services are being extended? No; it specifically leaves home delivery unchanged.

You need not memorize every word to answer well. You do need the relationships and conditions relevant to the question. This is the difference between simply covering text and selecting the information your purpose requires.

Use rereading to repair a specific gap

If the next sentence contradicts the meaning you formed, returning to an earlier clause can be the efficient choice. Try naming the gap before looking back: “What does ‘it’ refer to?” “Was that a cause or an association?” “Does the exception apply here?” A targeted return is easier to evaluate than repeatedly restarting the whole paragraph.

An experiment that masked words after the eyes passed them found poorer comprehension when useful rereading was unavailable. The finding supports preserving a way to repair understanding, while leaving open which returns are useful in a particular task. [3]

For a difficult section, make a short outline of the chain: claim, reason, example, qualification. If you cannot place a sentence in that chain, check whether it introduces a new topic or changes the previous claim. You are improving the representation of the argument, rather than trying to win a contest against the pause button.

Why a passage can feel easier than it is

Familiar wording can create confidence without demonstrating that you can explain the message. The problem becomes visible when a question asks you to connect ideas or supply an example that was not printed on the page.

Research on learning from screen and paper has examined predictions of performance alongside actual test results and study decisions. Such work makes a useful distinction between confidence and calibrated confidence. [4]

Try a simple check. Before looking at the answers, write how confident you are about each response. Then compare the explanation with the source. A confident error deserves attention: you may have missed a condition rather than merely forgotten a detail. A correct but uncertain explanation may need one more example to become secure.

Do this on material that matters to you. The purpose is to make your reading decisions better, not to attach a grade to every page you read.

Difficulty is not just a property of the page

A specialist may move quickly through material that is difficult for a beginner. The familiar reader brings meanings and relationships that the beginner is still constructing. Word-knowledge research places these language resources within the wider comprehension system. [5]

Difficulty and engagement can also interact with attention. In Soemer and Schiefele’s study, harder texts were associated with more mind wandering and lower interest. The study concerns those relationships; it is not evidence for a universal speed limit. [6]

If you repeatedly lose the thread, try a shorter section, look up the recurring terms or read a brief overview first. Keep the original question visible. A change in preparation may do more than pushing the display faster. For comparing progress, record unfamiliarity and difficulty so an easier passage does not masquerade as a training effect.

A practical routine for understanding that lasts

1. Before reading: name the purpose and inspect the structure. Decide which outcome you will check.
2. During reading: adjust pace where the argument changes. Mark a question rather than assuming an unclear sentence will resolve itself.
3. After a section: look away and explain the main point, supporting reason and important qualification.
4. Then verify: compare your explanation with the passage and correct a specific gap.
5. Later: return to the important idea and try to retrieve or apply it again.

Retrieval and distributed learning have evidence in memory and learning research. They should not be confused with a promise that every short exercise will transform reading speed. [7]

For a work document, your check might be a one-paragraph decision brief. For study, it might be explaining a process without the diagram. For leisure, you may simply want to follow the story and enjoy it. Choose the routine that fits the value of the reading.

Sources for this page

These links lead to the original paper, publisher or author archive. For annotations and limitations, visit the research library.

1. The role of knowledge in discourse comprehension: A construction-integration modelWalter Kintsch (1988). Theory and model
2. Conceptual processing of text during skimming and rapid sequential readingMichael Masson (1983). Experiments
3. Don’t Believe What You Read (Only Once): Comprehension Is Supported by Regressions During ReadingElizabeth Schotter, Randy Tran and Keith Rayner (2014). Experiment
4. Metacognitive regulation of text learning: On screen versus on paperRakefet Ackerman and Morris Goldsmith (2011). Experiments
5. Word Knowledge in a Theory of Reading ComprehensionCharles Perfetti and Joseph Stafura (2014). Theoretical synthesis
6. Text difficulty, topic interest, and mind wandering during readingAlexander Soemer and Ulrich Schiefele (2019). Experiment
7. Improving Students’ Learning With Effective Learning TechniquesJohn Dunlosky, Katherine Rawson, Elizabeth Marsh, Mitchell Nathan and Daniel Willingham (2013). Research review