What is metacognition in reading?

Metacognition is the awareness and regulation of your own cognitive processes — 'thinking about thinking.' In reading, metacognition refers to the conscious monitoring and control of your reading behaviour: knowing why you are reading a text, what you need to extract from it, when you are understanding versus merely decoding words, and whether your current reading strategy is serving your goal. Pressley and Afflerbach (1995) documented that expert readers are highly metacognitive, continuously adjusting speed, depth, and re-reading based on real-time comprehension monitoring.

What did the Klimovich 2023 study find?

Klimovich, Tiffin-Richards, and Richter (2023, DOI: 10.1111/1467-9817.12417) compared three groups: a commercial speed-reading app (n=10), a metacognitive training protocol (n=10), and a control (n=10). The metacognitive group received training only in setting a specific reading goal before reading. Both intervention groups produced significant speed gains with no comprehension loss; the metacognitive group's gains were statistically equivalent to the speed-reading app group. Eye-tracking data showed the metacognitive group made significantly fewer regression fixations — the mechanism for both speed and comprehension improvement.

How do reading regressions affect comprehension?

Regression fixations — backward eye movements to re-read earlier text — occur in all readers but are more frequent in inefficient reading. Rayner (1998) found that skilled readers make regressions on about 10–15% of fixations, compared to 20–30% in less skilled readers. Regressions are not always inefficient: they serve a legitimate function when genuine comprehension failure occurs. But regressions driven by distraction, uncertainty about goals, or low confidence — rather than actual comprehension failure — waste time without improving comprehension. Goal-setting reduces non-functional regressions.

What is a good reading intention to set?

A useful reading intention is specific and answerable: it names the information type you need and the decision it informs. Bad intention: 'understand this report.' Good intention: 'find the main recommendation, the budget figure, and the implementation timeline, so I can decide whether to escalate to the steering committee.' The specificity pre-loads your working memory with relevance criteria, which your visual system uses to flag matching content during the scan. Vague intentions produce vague scans.

Is metacognitive reading the same as active reading?

Metacognitive reading is a component of active reading, but not identical to it. Active reading encompasses annotation, questioning, summarising, and connecting — all behaviours that improve encoding and retention. Metacognition specifically refers to the monitoring and regulation layer: awareness of your own comprehension state and deliberate adjustment of strategy in response. You can annotate extensively (active reading) while remaining poorly calibrated about your comprehension (low metacognition). The Klimovich 2023 finding shows that the metacognitive component alone — just setting a goal — produces substantial speed gains independent of other active reading behaviours.

← Blog

Metacognitive reading: why knowing your goal makes you a faster reader

May 16, 20268 min readBy warpread.app

In 2023, a study by Klimovich, Tiffin-Richards, and Richter published in the Journal of Research in Reading posed a simple question: why does speed-reading training sometimes work? The researchers compared a commercial speed-reading application against a minimal metacognitive training protocol — participants in the metacognitive group were taught only to set a clear reading goal before each reading session.

Both intervention groups produced statistically significant speed gains with no comprehension loss. The metacognitive group's gains were equivalent to the speed-reading app group. The mechanism identified by eye-tracking: fewer regression fixations — backward re-reads — in the metacognitive group, producing faster passage times without reducing comprehension accuracy.

The implication is striking. A substantial portion of the speed advantage produced by commercial speed-reading training may be attributable to goal-setting — a skill that takes seconds to apply and no specialist software to practise.

What metacognition is and why it matters

Metacognition is the monitoring and regulation of your own cognitive processes. In reading, it is the awareness of how you are reading — whether you are comprehending or merely decoding words, whether your current pace matches the difficulty of the material, whether your reading strategy is serving your goal.

Pressley and Afflerbach (1995) documented the reading behaviour of expert readers through think-aloud protocols — asking skilled readers to narrate their reading process as it happened. Expert readers were highly metacognitive: they continuously set and updated their reading goals, monitored their comprehension in real time, adjusted their reading speed in response to difficulty, chose whether to re-read or continue, and connected new information to prior knowledge. Novice readers showed far less of this monitoring behaviour — they tended to read at a constant pace regardless of text difficulty and reported less awareness of comprehension failure.

The Klimovich et al. (2023) study operationalised the goal-setting component of this metacognitive behaviour and found it sufficient to produce measurable speed gains. Regulation of the other metacognitive variables — speed adjustment, re-reading decisions, connection to prior knowledge — presumably produces additional gains beyond goal-setting alone.

The regression problem

Rayner (1998, DOI: 10.1037/0033-2909.124.3.372) found that skilled readers regress on approximately 10–15% of fixations — backward eye movements to re-read earlier text. Less skilled readers regress on 20–30% of fixations. Not all regressions are wasteful: a regression after genuine comprehension failure (an unfamiliar term, a complex syntactic structure, an implicit reference) serves a legitimate comprehension function. But many regressions are non-functional — driven by distraction, low confidence, or uncertainty about what information matters.

Non-functional regressions are particularly common in readers who have not set a reading goal. Without a goal, the reader cannot distinguish between information that is relevant and information that is not, so uncertain comprehension of any sentence triggers a regression. A reader with a clear goal makes this distinction in real time: if a sentence is not relevant to the goal, incomplete comprehension of that sentence does not trigger a regression. The reader moves forward.

This is the mechanism Klimovich et al. (2023) identified. Goal-setting does not physically slow the eye down or speed it up. It recalibrates the comprehension-monitoring threshold — reducing non-functional regressions while preserving functional ones.

The three-question intention protocol

Before reading any document, answer three questions. This takes 30–60 seconds and is the minimal metacognitive preparation with the largest comprehension-per-minute return.

Question 1: What specifically am I trying to find out?

Name the information, not the topic. Not "I want to understand this report" but "I want to know the Q3 revenue figure, the gross margin trend, and the executive team's explanation for the miss." The specificity is load-bearing — vague goals produce vague attention filters.

Question 2: What decision or action does this information inform?

Connecting the reading goal to a downstream use creates motivational engagement that sustains attention across longer texts (Guthrie & Wigfield, 2000). It also refines the relevance criteria: if the decision is "whether to invest," then competitive risk sections are relevant and appendices about data methodology are not. If the decision is "how to critique this study," then methodology is central and the executive summary is peripheral.

Question 3: What do I already know about this topic?

Activating prior knowledge before reading primes the schema-integration process Kintsch (1988) describes. It takes 15–20 seconds. It produces measurably better comprehension of the subsequent text. It also flags gaps: if your answer to Question 3 reveals that you know very little about the topic, you should adjust your reading pace downward or plan for a survey pass before a full read.

Real-time monitoring during reading

Goal-setting before reading is the preparation step. The reading itself requires ongoing monitoring — the regulatory half of metacognition.

Monitor for goal-relevance: At each paragraph break, briefly test: did this paragraph advance my reading goal? If not, reduce speed for the next paragraph (it may be transitional) rather than maintaining a uniform pace that wastes attention on low-relevance content.

Monitor for comprehension versus decoding: There is a difference between understanding a sentence and processing its words. You can decode "the firm's EBITDA margin contracted by 340 basis points in Q3" without understanding what that implies about the firm's operational leverage. If you find yourself decoding without understanding, slow down or flag the passage — do not continue at skim speed through material that requires careful processing.

Monitor for diminishing returns: Long documents often have decreasing marginal information density — the first 30% carries most of the argument, the middle 40% provides evidence and examples, and the final 30% summarises and qualifies. Calibrate your pace to this structure: slower at the start (establishing the schema), faster through the evidence body (if the argument is already clear), careful again at the end (qualifications often matter).

Calibrating strategy to content

Klimovich et al.'s (2023) finding applies most strongly to non-fiction text with clear argumentative structure. For literary fiction, poetry, and legal documents — where sequential processing of every word is the point — metacognitive speed optimisation is counterproductive. See reading triage for the document-type decision framework.

For diagonal reading specifically, metacognitive goal-setting is the performance multiplier. The diagonal technique provides a systematic path through the text; the reading goal provides the relevance filter that makes each fixation count. Without a goal, diagonal reading can produce false confidence — you covered the text but captured noise as much as signal. With a goal, the spotted-pattern instinct (Pernice, 2017) activates naturally for content words matching your relevance criteria.

Try it now

The Diagonal Reader tool pairs naturally with the three-question protocol. Set your reading intention (30 seconds), then execute the diagonal scan with the tool's animated guide mode. The content-word highlighting makes goal-relevant terms immediately salient. Free, no account required.

The Diagonal Reading course covers metacognitive intention-setting in depth in Lesson 6, with evidence from Klimovich et al. (2023) and the Pressley & Afflerbach (1995) expert-reader research. The lesson includes a practice exercise with a dense business text and a three-question intention protocol. Six lessons, free, no account required.

References

Guthrie, J. T., & Wigfield, A. (2000). Engagement and motivation in reading. In M. L. Kamil, P. B. Mosenthal, P. D. Pearson, & R. Barr (Eds.), Handbook of Reading Research (Vol. 3, pp. 403–422). Erlbaum.
Kintsch, W. (1988). The role of knowledge in discourse comprehension. Psychological Review, 95(2), 163–182.
Klimovich, M., Tiffin-Richards, S. P., & Richter, T. (2023). Does speed-reading training work, and if so, why? Journal of Research in Reading, 46(2), 123–142. https://doi.org/10.1111/1467-9817.12417
Pernice, K. (2017). Text Scanning Patterns: Eyetracking Evidence. Nielsen Norman Group. https://www.nngroup.com/articles/text-scanning-patterns-eyetracking/
Pressley, M., & Afflerbach, P. (1995). Verbal Protocols of Reading: The Nature of Constructively Responsive Reading. Erlbaum.
Rayner, K. (1998). Eye movements in reading and information processing: 20 years of research. Psychological Bulletin, 124(3), 372–422. https://doi.org/10.1037/0033-2909.124.3.372

Topics

metacognitive readingmetacognition reading comprehensionreading goal settingreading intentionreading strategy researchhow to read more efficientlyreading comprehension strategiesspeed reading comprehension

Practice diagonal reading now

Paste any article into the Diagonal Reader to see the scan path in real time — or take the free 6-lesson course to learn the full technique with interactive exercises and quizzes.

Try the Diagonal Reader →Start the free course →

More on Diagonal Reading & Intelligent Skimming

What is diagonal reading? The evidence-based guide to structured skimming

Diagonal reading is a structured skimming technique that traces a staircase path through text, targeting content words rather than reading every line. Here's the science behind it and how to use it.

F-pattern reading: what eye-tracking research reveals about how we scan text

Eye-tracking studies with hundreds of participants reveal a universal F-shaped scanning pattern in readers. Here's what the research actually shows, what the variants mean, and how to design your reading around it.

How to skim read effectively without losing comprehension

Skimming is a learnable skill, not random fast reading. Here's a step-by-step evidence-based method for extracting key information at 2–3× your normal reading speed without losing comprehension.

The SQ3R method: why surveying before you read improves comprehension

SQ3R is the oldest research-validated reading system. Here's the evidence behind why its Survey step produces faster reading and better retention — and how to apply it to any document type.

← Previous

How to skim read effectively without losing comprehension