Highlighting shows low utility for improving learning outcomes¶

Claim · All claims
Evidence · 1 study · review r3 · q3 · n=multiple primary studies (narrative synthesis; individual samples span Air Force trainees, children and undergraduates)
Currency · newest study 2013 · not yet checked against its sources · not reviewed by a person

Highlighting and underlining text is one of the most widely used study strategies, yet it shows little benefit for learning outcomes relative to simply reading.

Subclaims¶

q3 i? A narrative review of highlighting/underlining research across children, undergraduates and Air Force trainees rates the technique as having low utility for learning, with little benefit over simply reading and a possible cost to inference-making. → Dunlosky et al. 2013

Evidence¶

Dunlosky et al. 2013¶

Dunlosky, J., Rawson, K. A., Marsh, E. J., Nathan, M. J., & Willingham, D. T. (2013). Improving students' learning with effective learning techniques: Promising directions from cognitive and educational psychology. Psychological Science in the Public Interest, 14(1), 4–58. doi:10.1177/1529100612453266

q3 · narrative review of primary experiments · i? · no pooled effect size reported · n=multiple primary studies (narrative synthesis; individual samples span Air Force trainees, children and undergraduates) · review · r3

A monograph by five cognitive/educational psychologists reviewing the empirical literature on ten study techniques, evaluated for generality across learning conditions, student characteristics, materials and criterion tasks. For highlighting/underlining specifically, the review synthesizes primary experiments (e.g., Fowler & Barker, 1974; Peterson, 1992) spanning Air Force basic trainees, children, remedial and typical undergraduates, using texts on topics from aerodynamics to enzymes, with immediate and delayed (up to two months) tests. Most studies found no benefit of highlighting over simply reading; one study (Peterson, 1992) found underlining actively hurt performance on inference questions requiring connections across the text, while leaving factual recall unaffected. The authors conclude the technique's benefit depends heavily on highlighting quality and prior knowledge, and that as typically practiced it provides little gain.

Discussion¶

Why highlighting underperforms. Highlighting is a passive, low-effort strategy: it can be performed fluently without the deep processing that produces durable memory [-M]. It may even create an illusion of competence — learners mistake the fluency of re-reading highlighted text for genuine understanding [-M], and highlighted fragments can disrupt the integration of ideas across a passage by isolating individual sentences from their context [-W]. This places highlighting in tension with cognitive load theory: rather than imposing the desirable difficulties that drive encoding, it minimizes effort at exactly the moment effort is needed.

Learner skill matters. The utility of highlighting depends heavily on how it is done [~M]. Learners who highlight indiscriminately — marking large proportions of the text — gain little, because over-highlighting eliminates the selectivity that might otherwise direct attention to key ideas [-M]. Even skilled highlighting mostly helps on questions that directly target the highlighted content, not on inference or integration questions [~M]. This mirrors the coherence principle: extraneous emphasis can divert attention from essential content rather than concentrating it [-M].

Better alternatives. Strategies that require generative processing — retrieval practice, self-explanation, and annotating with written notes — consistently outperform highlighting because they force learners to reconstruct and elaborate the material rather than merely select it [+M]. Highlighting may retain a modest role as a first-pass selection step that feeds a more demanding second pass (e.g., converting highlights into questions or summaries) [~W], but as a standalone strategy its expected payoff is low.

Open questions. Most evidence comes from short laboratory texts with immediate or short-delay tests; effects on complex, multi-session learning of authentic materials are less well established. Digital highlighting tools add features (colors, aggregation across documents) whose learning benefits remain largely untested.

Merged from “Highlighting Alone Is Ineffective” (highlighting-alone-is-ineffective): Why highlighting underperforms. Highlighting is a passive review behavior: learners can select text without connecting it to prior knowledge, elaborating on it, or self-testing on it. It also risks an illusion of competence — marked text feels familiar on rereading, which learners misinterpret as mastery. This places highlighting among the lower-yield study techniques in surveys of learning-strategy efficacy, in contrast to retrieval practice and elaborative techniques.

When marking can help. The picture is not uniformly negative. Marking may function as a first-pass selection step that supports later generative work — for example, when combined with annotating, margin notes, or self-explanation, or when learners are explicitly taught criteria for what deserves marking. Text-marking effects also interact with salience: making critical information visually distinct can guide attention, consistent with distinctiveness accounts of text marking (see Von Restorff effect and text marking). The claim here is therefore narrow: highlighting alone, as a complete study strategy, is ineffective — not that marking is harmful in all configurations.

Design implication. Where courses supply pre-highlighted or pre-annotated materials, designers should pair the marking with a task that forces processing — summarizing, questioning, or retrieval — rather than treating the highlight as the learning event. See the headings and highlight strategy for a structured variant.

Constraints on the claim. The ineffectiveness is conditional on highlighting serving as the sole study activity. Conditions under which marking fails or backfires include: learners with no criteria for relevance, who mark indiscriminately and then over-study marked text at the expense of unmarked content [-M]; rereading of highlighted text as a substitute for retrieval, which compounds the familiarity illusion [-M]; and pre-highlighted materials that impose the author's selection on learners without a task requiring them to process the selections [~M]. Where marking is paired with generative follow-up, the constraint does not apply.

Open questions. The review recorded above rates highlighting low utility from the primary experiments it summarizes, and reports no pooled effect size. Direct comparisons of highlighting with generative strategies are not separately recorded here.

Merged from “Highlighting Alone Low Utility” (highlighting-alone-low-utility): Why highlighting underperforms. Highlighting is a passive, low-effort strategy: it can be performed without deep comprehension of the text, and it does not require learners to retrieve, organize, or transform content. Learning benefits come from generative processing — summarizing, self-testing, annotating with one's own words — rather than from marking text itself. This aligns with the broader finding that annotating improves learning when annotation involves elaboration rather than mere marking.

Moderating conditions. Highlighting may serve a limited supporting role: it can direct attention during a second reading pass, or provide a scaffold for later generative work (e.g., turning highlights into flashcards or summaries). Its utility likely depends on what happens after highlighting, not the highlighting itself. Learners also frequently over-highlight, which can create an illusion of fluency — marked text feels familiar, encouraging premature cessation of study.

Boundary conditions. For learners with weak metacognitive monitoring, highlighting may actively mislead by privileging salient over important content — a concern related to the Von Restorff effect in text marking, where distinctiveness drives attention but not necessarily encoding. Strategy-training contexts should pair highlighting with explicit follow-up tasks rather than presenting it as a complete study method; structured approaches such as the headings and highlight strategy embed marking within a larger generative routine for this reason.

Open questions. Comparative evidence on highlighting versus alternative rereading strategies, and on whether highlighting quality (selectivity) predicts outcomes, still needs to be added to this page's evidence base.

Related Claims¶

• Annotating improves learning. — written annotation adds the generative processing that highlighting lacks
• Chunking reduces working memory load. — selective organization of material, unlike highlighting, restructures content for comprehension
• Coherence principle: irrelevant material hurts learning. — extraneous emphasis, like over-highlighting, can divert attention from essential content
• Cognitive overload degrades learning. — low-effort strategies fail because they do not engage the resources needed for durable encoding
• Headings and highlight strategy — a structured use of highlighting embedded in a larger strategy.
• Von Restorff effect — text marking — why salient, distinctive marks capture attention but do not guarantee encoding.
• In one experiment, students who generated their own underlining did not score significantly higher on tests than students given experimenter-generated underlining — related
• Relevancy of emphasized text directs attention and influences test performance — related
• Prior knowledge is not significantly related to test performance in this sample — related
• Summarization, highlighting, the keyword mnemonic, imagery for text, and rereading receive low utility assessments — a broader claim this one bears on
• Practice testing and distributed practice receive high utility assessments because their benefits generalize widely across learners, tasks, and educational contexts — related
• Ten-year-old readers predominantly underline subordinate passage details rather than superordinate conceptual material, even when conceptual postquestions focus attention on topic sentences — a narrower finding that bears on this claim
• Recall of passage details is most depressed when children receive both underlining and conceptual postquestions — a narrower finding that bears on this claim
• Across ten learning techniques, distributed practice and practice testing have the largest effects and underlining and summarization the smallest — related
• Underlining and adjunct questions may hinder rather than help recall in young readers, reflecting a comprehension rather than a metacomprehension deficit — a narrower finding that bears on this claim
• Engagement via annotation (underlining, highlighting, notes) shows no significant effect on comprehension — a narrower finding that bears on this claim
• Rereading Is A Low Utility Study Strategy — related
• Descriptive habit-strength ratings rank highlighting highest, followed by rereading, self-testing, and concept mapping — related
• Ineffective learning strategies show substantially higher habit strength than effective strategies in a proof-of-concept survey — a narrower finding that bears on this claim
• Anthropomorphic and colour features can capture and guide attention, overlapping with the signaling principle — related