Research into Practice: Highlighting, Distributed Study, and Metacognitive Beliefs

Research Overview and Context of Highlighting

• The study, titled "Highlighting and Its Relation to Distributed Study and Students’ Metacognitive Beliefs," was authored by Carole L. Yue, Benjamin C. Storm, Nate Kornell, and Elizabeth Ligon Bjork and published in Educational Psychology Review in July $2014$.

• Highlighting and underlining are described as ubiquitous practices among students, often visible in used textbooks in college bookstores.

• Students generally believe that marking text helps them remember information better or makes future study sessions more effective.

• Despite its prevalence, academic evidence regarding the benefits of highlighting is mixed:

  • Several studies have shown significant benefits for underlined or highlighted text (e.g., Fass and Schumacher $1978$; Fowler and Barker $1974$; Nist and Hogrebe $1987$; Nist and Simpson $1988$; Johnson $1988$; Rickards and August $1975$).

  • Other studies have failed to show such benefits (e.g., Arnold $1942$; Hoon $1974$; Idstein and Jenkins $1972$; Peterson $1992$; Stordahl and Christensen $1956$; Wade and Trathen $1989$).

Potential Advantages of Highlighting as a Study Strategy

• Depth-of-Processing Perspective:

  • Drawing from Craik and Lockhart ($1972$), the act of deciding what to highlight and what to ignore may lead students to process information at a deeper, more evaluative level.

  • Research suggests learner-generated highlighting produces better test performance than experimenter-generated highlighting (Fowler and Barker $1974$; Rickards and August $1975$).

  • Training students to read a paragraph and decide what is conceptually important before highlighting improves performance compared to those without training (Leutner et al. $2007$).

• The von Restorff Effect:

  • This phenomenon (Wallace $1965$) suggests that marked text becomes more memorable because it stands out from the surrounding non-highlighted text.

  • Evidence shows that students reading pre-highlighted passages recall more of the highlighted and less of the non-highlighted information (Fowler and Barker $1974$; Silvers and Kreiner $1997$).

• Encoding Variability:

  • Varying the context or processes of repeated learning facilitates retention and transfer (Smith et al. $1978$).

  • By selectively marking text, learners change the visual nature of the text; when re-reading, they may encode this modified text in a new way, increasing the likelihood of successful retrieval in future contexts.

Potential Disadvantages and Metacognitive Barriers

• Ineffective Application:

  • Highlighting may be ineffective or detrimental if students do not know how to do it effectively, using it merely to track progress rather than for deep processing (Bell and Limber $2010$; Stordahl and Christensen $1956$).

• Interference with Established Habits:

  • Forcing students who do not usually highlight to use the technique may interfere with their natural, effective encoding techniques (Howe and Singer $1975$).

• Illusion of Knowing and Competence:

  • Students who rely heavily on highlighting may suffer from an "illusion of knowing" (Bjork $1999$, $2013$; Koriat and Bjork $2005$).

  • Re-reading highlighted text can create a sense of "processing fluency" where the student assumes they have mastered the material simply because it is marked and looks familiar.

  • This leads to a shallower processing during re-study, preventing effective long-term retention (Peterson $1992$).

Highlighting and the Spacing Effect

• Prevalence of Re-study: Approximately $60\%$ of undergraduates report using marked passages as a guide for later restudy (Kornell and Bjork $2007$).

• Distributed Study (Spacing): Often called a "desirable difficulty" (Bjork $1994$; Bjork and Bjork $2011$), spacing typically results in greater long-term retention despite feeling more difficult during encoding.

• Specific Hypotheses Regarding Highlighting and Spacing:

  • Highlighting might actually be more beneficial in massed conditions (immediate re-reading) than in spaced conditions.

  • Massed study usually suffers from low encoding variability; highlighting might attenuate this by forcing a different encoding during the second reading.

  • Highlighting may dispel the sense of fluency that usually discourages deep processing when reading the same text back-to-back.

Experimental Methodology

• Participants: $184$ UCLA undergraduates ($M_{age} = 19.9$) participating for course credit.

• Materials:

  • An $856$-word passage about ground water from the U.S. Geological Survey website.

  • $12$ critical phrases containing key keywords (e.g., the keyword "recharge" in the sentence: "Water seeping down from the land surface adds to the ground water and is called recharge water.").

• Design: A $2 \times 2$ between-subjects design:

  1. Highlighting vs. No-Highlighting.

  2. Massed re-reading (immediate) vs. Spaced re-reading ($30$-minute interval).

• Procedure:

  • Initial reading: $6$ minutes.

  • Highlighting group: Used standard yellow highlighters as they typically would.

  • Massed group: Re-read immediately.

  • Spaced group: $30$-minute distractor activity before re-reading.

  • All participants took a questionnaire on metacognitive beliefs after the second reading.

  • Test: A fill-in-the-blank test administered after a $1$-week delay.

Results: Effects of Highlighting and Spacing on Learning

• Retention Benefits:

  • A robust and significant benefit of highlighting was found in the massed condition ($M = 0.12$ benefit), $t(90) = 2.89$, $p < 0.01$, $d = 0.60$.

  • A non-significant benefit of highlighting was observed in the spaced condition ($M = 0.04$ benefit), $t(90) = 0.92$, $p = 0.36$, $d = 0.19$.

• ANOVA Main Effects:

  • Main effect of highlighting: Highlighters ($M = 0.40$, $SE = 0.02$) outperformed non-highlighters ($M = 0.32$, $SE = 0.02$), $F(1, 180) = 7.22$, $MSE = 0.57$, $p < 0.01$.

  • Spacing main effect: Not significant ($F < 1$), with massed ($M = 0.37$) vs spaced ($M = 0.36$). This is attributed to the length of the passage causing inherent spacing between key phrases even in massed conditions.

Results: Individual Differences and Highlighting Behavior

• Highlighting Activity:

  • Participants highlighted an average of $191.9$ words ($SE = 13.5$, $Mdn = 175$), which was $22.4\%$ of the passage.

  • Highlighting was efficient: $72.5\%$ of keywords ($8.7$ out of $12$) were captured in the highlighted areas.

• Highlighted vs. Non-highlighted Material:

  • Highlighted information was recalled better ($M = 0.44$) than non-highlighted information ($M = 0.30$), $t(80) = 3.67$, $p < 0.01$, $d = 0.50$.

  • Importantly, highlighting did not impair recall of non-highlighted info compared to the control group ($M = 0.32$), which contradicts the von Restorff explanation.

• Heavy vs. Light Highlighters (Median Split):

  • Heavy highlighters ($M = 282.8$ words) highlighted significantly more total words than light highlighters ($M = 101.0$ words).

  • Heavy highlighters captured more keywords ($9.7$) than light highlighters ($7.8$).

  • However, light highlighters numerically outperformed heavy highlighters ($0.41$ vs $0.39$) on the final test.

  • Efficiency Score (Keywords/Total Words): Light highlighters were significantly more efficient ($M = 0.08$) than heavy highlighters ($M = 0.03$), $t(89) = 7.20$, $p < 0.01$, $d = 1.50$.

Results: Metacognitive Beliefs and Group Classification

• Belief Survey: Participants rated "I feel that highlighters are an important part of my studying" on a $1$-$9$ scale.

  • Pro-highlighters ($\text{rating} \ge 7$, $n = 74$): Mean activity $219.9$ words. Performance: Highlighters ($0.30$), Non-highlighters ($0.30$). Benefit: $0\%$.

  • Unsure ($\text{rating } 4$-$6$, $n = 55$): Mean activity $212.2$ words. Performance: Highlighters ($0.45$), Non-highlighters ($0.28$). Benefit: $17\%$.

  • Anti-highlighters ($\text{rating} \le 3$, $n = 55$): Mean activity $132.1$ words. Performance: Highlighters ($0.50$), Non-highlighters ($0.40$). Benefit: $10\%$.

• The Paradox of Belief:

  • Anti-highlighters ($M = 0.45$) significantly outperformed unsure participants ($M = 0.37$), who outperformed pro-highlighters ($M = 0.30$).

  • Those who valued highlighting the most (pro-highlighters) profited the least ($0\%$ benefit).

Additional Manipulations: The Effect of Testing

• A separate manipulation explored the interaction of the "testing effect" (Roediger and Karpicke $2006$) with highlighting.

• Participants received an immediate test on $6$ of the $12$ keywords shortly after the second reading.

• Results showed a large benefit for immediate testing: keywords tested immediately were remembered significantly better on the delayed test ($M = 0.47$, $SE = 0.02$) than those not tested immediately ($M = 0.26$, $SE = 0.02$), $F(1, 180) = 102.99$, $MSE = 4.24$, $p < 0.001$.

• However, the effect of testing did not interact with spacing ($p = 0.83$) or highlighting ($p = 0.33$).

General Discussion and Educational Implications

• Highlighting as a Desirable Difficulty: For students unaccustomed to highlighting, the act might force them to think about and process text differently, leading to better memory. This explains why anti-highlighters benefited more than pro-highlighters.

• Dispelling Fluency: Highlighting appears particularly beneficial in massed study because it may serve to dispel the misleading sense of fluency that prevents deep processing during back-to-back readings.

• Refutation of simple von Restorff Effect: Because non-highlighted information was not impaired, highlighting does not merely shift attention but improves the conceptual mapping/processing of the passage.

• Training Recommendations:

  • Instruction in effective highlighting is described as "obligatory" rather than just warranted, given that re-reading is a top-rated study activity among students.

  • Training should encourage students to justify their highlight choices and ask questions about that choice during re-reading.

• Final Conclusion: Highlighting per se is not beneficial; its value lies in how it changes the way students read and think about text. Relying on the act itself (as seen in heavy highlighters) without corresponding cognitive effort results in an "illusion of competence."