ArticlePDF Available

Abstract and Figures

Use of highlighting is a prevalent study strategy among students, but evidence regarding its benefit for learning is mixed. We examined highlighting in relation to distributed study and students’ attitudes about highlighting as a study strategy. Participants read a text passage twice while highlighting or not, with their readings either distributed or massed, and followed by a week-delayed test. An overall benefit of highlighting occurred, with highlighting being especially beneficial with massed readings of the passages. Importantly, highlighting did not impair knowledge of non-highlighted information. Interestingly, those students reporting that they did not think highlighting was beneficial or were unsure about its benefits actually benefitted more from highlighting than did students who were pro-highlighting. Overall, our results indicate that under some conditions, highlighting can be a beneficial study strategy for learning and argue for students being trained in how to optimize the potential benefits of their highlighting behavior.
Content may be subject to copyright.
Highlighting and Its Relation to Distributed Study
and StudentsMetacognitive Beliefs
Carole L. Yue &Benjamin C. Storm &Nate Kornell &
Elizabeth Ligon Bjork
Published online: 22 July 2014
#Springer Science+Business Media New York 2014
Abstract Use of highlighting is a prevalent study strategy among students, but evidence
regarding its benefit for learning is mixed. We examined highlighting in relation to distributed
study and studentsattitudes about highlighting as a study strategy. Participants read a text
passage twice while highlighting or not, with their readings either distributed or massed, and
followed by a week-delayed test. An overall benefit of highlighting occurred, with highlighting
being especially beneficial with massed readings of the passages. Importantly, highlighting did
not impair knowledge of non-highlighted information. Interestingly, those students reporting
that they did not think highlighting was beneficial or were unsure about its benefits actually
benefitted more from highlighting than did students who were pro-highlighting. Overall, our
results indicate that under some conditions, highlighting can be a beneficial study strategy for
learning and argue for students being trained in how to optimize the potential benefits of their
highlighting behavior.
Keywords Highlighting .Spacing .Text marking .Metacognitive beliefs about study strategies
One needs only to browse through used textbooks in a college bookstore to see that text-
marking, either by highlighting or underlining, is a ubiquitous practice among students, with
many believing that marking text will help them remember the selected information better or
make a later study session more effective. Whether text-marking actually does benefit later
recall, however, is debatable: Several studies have shown a significant benefit for underlined or
highlighted text (e.g., Fass and Schumacher 1978; Fowler and Barker 1974; Nist and Hogrebe
Educ Psychol Rev (2015) 27:6978
DOI 10.1007/s10648-014-9277-z
C. L. Yue (*)
Covenant College, 14049 Lookout Mountain, GA 30750, USA
B. C. Storm
University of California, Santa Cruz, Santa Cruz, CA, USA
N. Kornell
Williams College, Williamston, MA, USA
E. L. Bjork
University of California, Los Angeles, Los Angeles, CA, USA
Author's personal copy
1987; Nist and Simpson 1988; Johnson 1988; Rickards and August 1975), whereas others
have not (e.g., Arnold 1942; Hoon 1974; Idstein and Jenkins 1972; Peterson 1992; Stordahl
and Christensen 1956; Wade and Trathen 1989). Given the prevalence of highlighting as a
study technique among students, however, learning more about the circumstances under which
it might be (or could be made to be) a more effective strategy seems a worthwhile goal from an
educational standpoint, particularly if such research might reveal guidelines that could be
given to students regarding how to make highlighting more beneficial for their learning.
Potential Advantages of Highlighting
There are several reasons to expect text-marking to benefit learning. From a depth-of-processing
perspective, just the act of deciding what to mark and what not to mark may lead students to
process textual information at a deeper, more evaluative level than they would when simply
reading it (Craik and Lockhart 1972;NistandHogrebe1987). Consistent with this idea, learner-
generated highlighting tends to produce better test performance than experimenter-generated
highlighting (Fowler and Barker 1974;RickardsandAugust1975;RickardsandDenner1979;
but see Nist and Hogrebe 1987). Additionally, when students are trained in highlighting tech-
niques (i.e., to read a paragraph, decide what is conceptually important, and then highlight that
information), they perform better than studentswhodonotreceivesuchtraining(Leutneretal.
2007), indicating that appropriate cognitive activity during highlighting can enhance its benefits.
Another potential benefit of text-marking could be a type of von Restorff effect (Wallace 1965).
Specifically, highlighting may make the marked portion of text more memorable because it stands
out from the surrounding non-highlighted text. Indeed, some evidence supports this type of role for
highlighting: When students read pre-highlighted passages, they recall more of the highlighted
information and less of the non-highlighted information compared to students who receive an
unmarked copy of the same passage (Fowler and Barker 1974;SilversandKreiner1997).
Highlighting might also enhance the effectiveness of re-study opportunities via encoding
variability. Varying the context or particular processes involved in repeated learning opportunities
has been found to facilitate performance on later tests of retention and transfer, the explanation
being that learning is less likely to become contextualized under such circumstances (e.g., Smith
et al. 1978). Variability is presumed to be effective because it increases the likelihood that
participants will encode to-be-learned information in slightly different ways, thus increasing their
ability to retrieve that information when tested in another context in the future. By selectively
marking text, learners change the text as they read it; consequently, when re-reading marked text,
learners may read and encode that text in a new way, thereby making it more memorable.
Potential Disadvantages of Highlighting
In contrast to these arguments, others have argued that selectively highlighting text might be
ineffective or even detrimental to learning (Dunlosky et al. 2013; Idstein and Jenkins 1972;
Peterson 1992; Stordahl and Christensen 1956). One argument is that students often do not
know how to highlight effectively, so such activity primarily amounts to a mechanism for
tracking progress and does not involve deeper processing (Stordahl and Christensen 1956; Bell
and Limber 2010). Another relevant factor is whether students are accustomed to using a
highlighter (Brown and Smiley 1978). Forcing readers who never use highlighters to do so
may interfere with their learning and prevent them from employing the type of encoding
techniques they usually find beneficial (Howe and Singer 1975).
70 Educ Psychol Rev (2015) 27:6978
Author's personal copy
Additionally, studentsmetacognitive beliefs about highlighting may limit its effec-
tiveness as a learning tool. Students who rely on highlighters and think they are
particularly effective, for example, may suffer from an illusion of knowing or compe-
tence (Bjork 1999,2013;KoriatandBjork2005). Specifically, such students may
process highlighted material in a less meaningful way when re-reading than if that
material were not highlighted. While re-reading, such students may only quickly glance
over highlighted text, incorrectly assuming that because they have already highlighted
that information, it is deeply encoded in memory, a misbelief that is probably supported
by the apparent processing fluency that learners would experience during such re-
reading. In this way, highlighting could ironically impair memory for critical informa-
tion by preventing students from restudying the information in a way that effectively
promotes long-term retention (cf. Peterson 1992).
Highlighting and the Distribution of Study
In evaluating the efficacy of text-marking, it is important to consider that students often
mark text for the purpose of guiding their future study. For example, in a survey of 472
undergraduates, 60 % reported using marked passages as a guide for later restudy
(Kornell and Bjork 2007). Thus, it seems critical to examine how text-marking might
interact with the spacing of study activities. Distributed study, or spacing, is a desirable
difficulty in that it typically results in greater long-term retention even though it can make
learning feel more difficult during encoding (Bjork 1994; Bjork and Bjork 2011). Indeed,
spaced study of educationally relevant materials has been repeatedly shown to improve
retention compared to massed study (e.g., Dempster 1996;Kornell2009; Sobel et al.
Highlighting, however, might actually be more beneficial in massed conditions than
in spaced conditions. Massed study is often presumed to be inferior to spaced study
because it involves less encoding variability and because it limits the effectiveness of
the second study opportunity (Hintzman 1974). If highlighting attenuates these disad-
vantages by leading learners to encode the passage differently in a second reading, such
beneficial effects should be relatively greater in massed than spaced conditions. More-
over, active highlighting might possibly dispel the misleading effects of fluency that
tend to discourage deep processing of information upon re-reading. If so, because the
sense of fluency would be stronger the closer in time the second reading follows the
first, such an effect of highlighting should be more beneficial the sooner the second
reading follows the first.
Overview of the Present Study
The goals of the present research were to assess possible benefits of highlighting as well as
individual differences in the use of highlighting and to explore effects of highlighting in
relation to distributed study and metacognitive beliefs about highlighting as a study tool. To
this end, we asked students to study a passage twice, either massed (i.e., back-to-back with no
separation between study opportunities) or spaced (i.e., successive study opportunities sepa-
rated by a 30-min interval), with half studying the passages without using a highlighter and
half studying the passage using a highlighter. All participants then took a test after a 1-week
delay. We also collected data on studentshighlighting preferences.
Educ Psychol Rev (2015) 27:6978 71
Author's personal copy
A total of 184 UCLA undergraduates (M
=19.9) participated for partial credit in a psychol-
ogy course.
Participants read a passage about ground water (856 words) from the U.S. Geological Survey
website. Twelve critical phrases, each containing a different keyword, were selected from the
passage (e.g., the term recharge was the keyword in the phrase: Water seeping down from the
land surface adds to the ground water and is called recharge water.). Then, 12 fill-in-the-
blank questions were created from these phrases by deleting the keyword and asking partic-
ipants to provide it on the final test (e.g., Water seeping down from the land surface adds to the
ground water and is called __________ water).
Design and Procedure
Participants were randomly assigned to either the highlighting or no-highlighting conditions
and to either the massed or spaced re-reading conditions; thus, our design employed two
between-subject variables: highlighting and spacing.
Upon arrival, participants were seated alone at a desk and asked to read the passage in its
entirety, which they were given 6 min to do. Participants in the highlighting condition received
a standard yellow highlighter and told to use it however they typically would while studying
material for a class. Participants in the no-highlighting condition were not given a highlighter
and were simply instructed to read the passage as though they were studying material for a
After the initial reading, participants in the massed condition were immediately asked to
study the passage again, while those in the spaced condition participated in a 30-min unrelated
distractor activity before re-studying the passage. In the highlighting condition, participants re-
read their previously highlighted passage (with their markings still there) and were again told
to use the highlighter however they typically would while studying for a class. Participants in
the no-highlighting condition were simply instructed to re-read the passage as though they
were studying for a class.
After the second reading, all participants were given a brief questionnaire asking them to
indicate the extent to which they either agreed or disagreed with a set of statements exploring
their metacognitive beliefs about learning and study strategies, such as, I feel that highlighters
As a separate manipulation, we also explored whether the benefits of testing (Roediger and Karpicke 2006)
might interact with highlighting. Specifically, participants were given an immediate test on six of the twelve fill-
in-the-blank questions shortly after the second reading of the passage. All twelve questions were then tested after
the 1-week delay, allowing us to assess the benefits of the earlier test. Although we observed a large benefit of
testing, F(1, 180)=102.99, MSE= 4.24, p<0.001, with keywords tested immediately remembered significantly
better on the delayed test (M=0.47, SE=0.02) than were keywords not tested immediately (M=0.26, SE=0.02),
the effect of testing did not interact with either the spacing (p=0.83) or highlighting (p=0.33) manipulations.
Consequently, for the sake of succinctness, and because educators are most likely to be interested in how
highlighting affects long-term learning and performance, we collapsed all data from the tested versus non-tested
conditions and report only one score to reflect the week-delayed final recall performance.
72 Educ Psychol Rev (2015) 27:6978
Author's personal copy
are an important part of my studying.One week later, all participants were given the fill-in-
the-blank test for the 12 critical phrases from the passage.
Results and Discussion
How Do Highlighting and Spacing Affect Learning?
Average correct performance obtained on the final fill-in-the-blank test in each of our four
conditions is illustrated in Fig. 1, and as indicated there, one of our pre-study conjecturesthat
highlighting might be more beneficial in massed than spaced conditionsdid receive some
support. Planned-comparison t-tests revealed that whereas a nonsignificant benefit of
highlighting was observed in the spaced condition (M=0.04 benefit), t(90)=0.92, p=0.36,
d=0.19, a robust and significant benefit of highlighting was observed in the massed condition
(M=0.12 benefit), t(90)=2.89, p<0.01, d=0.60.
Additionally, we performed an overall analysis on our data using a 2(spaced vs. massed)×
2(highlighting vs. no-highlighting) between-subjects ANOVA. Not surprisingly, given the
pattern of results shown in Fig. 1, there was no main effect of spacing, F(1, 180) <1, MSE=
0.03, with performance averaged across the two massed conditions (M=0.37, SE=0.02) not
differing from that averaged across the two spaced condition (M=0.36, SE= 0.02); but, there
was a significant main effect of highlighting, with the average performance of highlighters
(M=0.40; SE=0.02) being significantly better than the average performance of non-
highlighters (M=0.32; SE= 0.02), F(1, 180)= 7.22, MSE= 0.57, p<0.01. The interaction be-
tween highlighting and spacing, however, did not reach statistical significance, F(1, 180)=
1.93, MSE=0.15, p=0.17.
Given the typical robustness of the spacing effect that we did not find a benefit of spacing
even for the non-highlighters would suggest that our particular spacing manipulation was not
sufficiently strong. Indeed, because our to-be-leaned material was a three-page text passage
requiring up to 6 min to read once, a participants re-encountering of given key phrases would
have been spaced by several minutes even in our massed condition. Thus, the interval between
spaced encounters of the same key phrases may have not been sufficiently increased in our
spacing condition versus our massed condition to allow a spacing benefit to emerge.
Proportion of Keywords Recalled on
Final Test
Fig. 1 Final test performance by spacing and highlighting conditions
Educ Psychol Rev (2015) 27:6978 73
Author's personal copy
How Do Individual Differences in Highlighting Behavior Affect Learning?
Overall, participants highlighted an average of 191.9 (SE=13.5) words (Mdn =175), and
their highlighting behavior was quite efficient. Despite participants highlighting only
22.4 % of the passage, an average of 8.7 (SE=0.27) or 72.5 % of the 12 critical
keywords were included in their highlighting. Eighty-one participants highlighted some
keywords but not others, allowing for within-subjects comparisons of recall for
highlighted versus non-highlighted keywords. We first examined whether highlighted
information was better recalled than non-highlighted information, and importantly, it
was. Participants were more likely to answer a test question correctly if they had
highlighted information relevant to that question (M=0.44) than if they had not (M=
0.30), t(80)=3.67, p<0.01, d=0.50. Furthermore, recall performance of highlighting
participants for non-highlighted material was no different from the recall performance
of non-highlighting participants (M=0.32), t(174)= 0.56, p=0.58. Thus, highlighting
appears to improve the retention of highlighted material without significant cost to the
retention of the non-highlighted materiala finding inconsistent with the von Restorff
To explore a possible relationship between highlighting activity and later recall, we
conducted a median-split analysis separating participants into heavy highlighters and light
highlighters. As shown in Table 1, heavy highlighters marked significantly more words than
light highlighters, t(90)= 9.41, p<0.001, d=1.96, including more keywords, t(90) =3.81,
p<0.001, d=0.97. Importantly, however, heavy highlighters did not outperform light
highlighters at final test. If anything, light highlighters numerically outperformed heavy
highlighters, suggesting that the benefits of highlighting do not stem from the mere act of
highlighting alone.
Possibly, the light highlighters put more cognitive effort and analysis into deciding what to
highlight, resulting in fewer highlighted words, but deeper processing of those words com-
pared to words highlighted by heavy highlighters. We explored this conjecture by calculating
an efficiency score: the number of keywords highlighted divided by the total number of words
highlighted. By this measure, light highlighters were significantly more efficient (M=0.08,
SE= 0.01) than heavy highlighters (M=0.03, SE= 0.01), t(89)= 7.20, p<0.01, d=1.50. Thus, it
would appear that light highlighters were more selective in their highlighting than heavy
highlighters, perhaps reflecting more cognitive effort being given to their highlighting
How do StudentsBeliefs about Highlighting Relate to the Benefits of Highlighting?
According to the questionnaire responses, many students use highlighters and believe them to
be an important component of their studying. When asked to rate the statement, I typically
read my text books while using a highlighteron a scale from 19with 1 meaning
completely disagree,5 meaning unsure,and 9 meaning completely agree”—48 % of
Table 1 Highlighting activity and final test performance by highlighting-classification group
Total words highlighted
Keywords highlighted
Final test performance
Heavy highlighters 282.8 (17.7) 9.7 (0.3) 0.39 (0.03)
Light highlighters 101.0 (7.8) 7.8 (0.4) 0.41 (0.03)
74 Educ Psychol Rev (2015) 27:6978
Author's personal copy
the participants selected a 7 or above (M=5.5; SD= 2.8). Furthermore, when asked to rate the
statement, I feel that highlighters are an important part of my studyingusing the same 1-9
scale, 41 % of the participants selected a 7 or above (M=5.3; SD=2.6). If anything, these
ratings probably underestimate student text-marking behavior, as participants may have
restricted their responses to highlighting, discounting similar text-marking activities such as
To see whether differences in opinions about highlighting predicted differences in
highlighting activity or recall performance, we separated participants into three groups: pro-
highlighters, unsure, and anti-highlighters. Pro-highlighters were the 74 participants rating the
statement, I feel that highlighters are an important part of my studyingwith a 7 or above;
unsure participants were the 55 participants rating the statement between 4 and 6, and anti-
highlighters were the 55 participants rating the statement 3 or below. Unsurprisingly, a main
effect of group was observed on highlighting activity, F(2,89)=4.39, MSE =68,812.20,
p<0.05. As shown in Table 2, pro-highlighters and those who were unsure highlighted
significantly more words than did anti-highlighters, t(63)= 3.35, p<0.01, d=0.86, t(52)=
2.19, p= 0.03, d=0.60, respectively, and a similar pattern was observed for keywords
To see if opinions about the importance of highlighting were related to retention, we
conducted a 2(highlighting vs. no-highlighting)×2(spaced vs. massed)×3(pro-highlighters
vs. unsure vs. anti-highlighters) between-subjects ANOVA on the final test scores. Spacing
did not interact significantly with any variable, so we collapsed across the massed and spaced
conditions. Interestingly, a significant effect of group emerged, such that anti-highlighters (M=
0.45; SE=0.03) outperformed unsure participants (M=0.37; SE= 0.03), who outperformed
pro-highlighters (M=0.30; SE=0.02), F(2, 172)=9.60, MSE=0.34, p<0.001, with individual
t-tests confirming each of the between-group differences to be statistically significant, average
Although anti-highlighters outperformed pro-highlighters, we nonetheless expected
pro-highlighters to benefit most from being allowed to use a highlighter. As shown in
Tab le 2, however, the opposite was observed. Anti-highlighters benefited marginally from
use of a highlighter (M=10 % benefit), t(53)=1.69, p=0.09, d=0.50, and unsure partic-
ipants benefited significantly (M=17 % benefit), t(53)=3.92, p<0.001, d= 1.00), but pro-
highlighters did not benefit at all (M= 0 % benefit), t(72) <1. The interaction between
highlighting group and highlighting condition was statistically significant, F(2, 172)=
3.90, MSE=0.14, p=0.02.
Table 2 Highlighting activity and final test performance by highlighting efficacy and belief classification
Experimental condition and belief
Total words
highlighted (SE)
highlighted (SE)
Final test performance
Pro-highlighters 219.9 (20.3) 8.9 (.4) 0.30 (0.03)
Unsure highlighters 212.2 (24.1) 7.6 (.6) 0.45 (0.03)
Anti-highlighters 132.1 (24.1) 9.6 (.3) 0.50 (0.04)
Pro-highlighters ––0.30 (0.03)
Unsure highlighters ––0.28 (0.03)
Anti-highlighters ––0.40 (0.04)
Educ Psychol Rev (2015) 27:6978 75
Author's personal copy
General Discussion
The goal of the present research was to explore potential benefits of highlighting in relation to
distributed study and studentsmetacognitive beliefs about highlighting as a study tool. We
found that highlighting improved later cued recall of highlighted information without
impairing recall of non-highlighted information from a text passage, a finding that is incon-
sistent with a von Restorff-based explanation. That is, highlighting did not seem to enjoy its
benefit merely by making highlighted text stand out upon re-study. Furthermore, we found that
the benefit of highlighting was numerically greater when participants read the passage twice
without delay, suggesting that highlighting may be particularly beneficial when students re-
read text passages immediately.
The results of the present research suggest that highlighting, far from being an ineffective
study technique (Dunlosky et al. 2013), can facilitate long-term retentionparticularly when
students, possibly owing to limited available study time, engage in massed re-readings or study
sessions. In such situations, students could probably improve the effectiveness of their study
via selective highlighting because such a practice would lead them to think about why they
initially selected certain words or phrases to highlight, resulting in deeper processing during
subsequent readings.
If initial highlighting does encourage learners to engage in such considerations about
previously highlighted material, then highlighting might have some of its beneficial ef-
fectsas suggested earlierby serving to dispel the misleading effect of fluency arising
during a subsequent re-reading. Indeed, the sense of fluency typically felt during an immediate
re-reading has been suggested as a major factor in why two back-to-back readings of a chapter
result in no better learning than just one reading (Callender and McDaniel 2009). Accordingly,
times when a sense of fluency is high and most likely to discourage deep processing on a
second reading should also be the times when having previously highlighted the passage
would be most beneficiala pattern consistent with our finding of a greater benefit for
highlighting when text readings were massed versus spaced.
A surprising finding of the present study is that participants who valued highlighters the
most profited least from their use. One possible reason for this finding is that participants who
were unaccustomed to highlighting put more effort into the act of highlighting, with the
ultimate result of better retention. From this perspective, highlighting could be characterized
as a desirable difficulty, at least for some students, because it forces them to think about and
process text differently than they typically would and in a way that ultimately leads to better
memory for that text. These results also suggest that even if participants were prevented from
engaging in the type of study processes they normally employ, the costs of such prevention did
not outweigh the benefits of using a highlighter.
Our results also indicate that training students how to highlight effectively could help
promote useful study strategies. Students often re-read text passages as a study activity, and
indeed, many rate it as their no. 1 study activity (Dunlosky et al. 2013; Karpicke et al. 2009;
Kornell and Bjork 2007,2009). Presumably, then, students are highly likely to persist in this
activity. Accordingly, instructing them how to do so optimally would seem not just warranted,
but obligatory. Highlighting training such as that proposed by Leutner et al. (2007), for
example, could well be helpful in furthering this goal, evenor perhaps especiallyfor those
who already believe that highlighting is a beneficial study technique. Such training should
involve encouraging students to think carefully about which sections of the text should be
highlighted and to justify their choices, as well as asking those questions again when re-
reading a previously highlighted section. Such questioning during highlighting and re-reading
should evoke two beneficial activities for improved retention: deeper processing and retrieval
76 Educ Psychol Rev (2015) 27:6978
Author's personal copy
practice, both of which have been repeatedly shown to improve retention (e.g., Craik and
Lockhart 1972; Roediger and Karpicke 2006).
More generally, the present work provides another example of what Bjork (1999) and
others (e.g., Koriat and Bjork 2005) have referred to as an illusion of competence. Specifically,
learners can be fooled by objective and subjective indices of performance into thinking that a
given manipulation is useful for learning even when it is not. In this context, individuals who
become reliant on highlighters for studyingsuch as the pro-highlighters or heavy
highlighters in the present studymay think that the act of highlighting is helpful in and of
itself. As the present results confirm, however, simply the act of highlighting text is not
sufficient to promote its retention. Indeed, despite the fact that highlighting a relevant portion
of a text was clearly beneficial, more overall highlighting activity tended to lead to worsenot
betterperformance at final test. Clearly, it is not highlighting per se that is beneficial; rather,
it is how highlighting changes the way students read and think about text that is beneficial.
Acknowledgments We thank John Nestojko and Gabriela Pocasangre for their contributions to this project.
This research was supported by Grant 29192G from the James S. McDonnell Foundation.
Arnold, M. B. (1942). A study of the relation to breakdown. Journal of Educational Psychology, 33,449457.
Bell, K. E., & Limber, J. E. (2010). Reading skill, textbook marking, and course performance. Literacy Research
and Instruction, 49, 5667. doi:10.1080/19388070802695879.
Bjork, R. A. (1994). Institutional impediments to effective training. In D. Druckman & R. A. Bjork (Eds.),
Learning, remembering, believing: enhancing human performance (pp. 295306). Washington, DC:
National Academy Press.
Bjork, R. A. (1999). Assessing our own competence: heuristics and illusions. In D. Gopher & A. Koriat (Eds.),
Attention and performance XVII. Cognitive regulation of performance: interaction of theory and application
(pp. 435459). Cambridge: MIT Press.
Bjork, E. L., & Bjork, R. A. (2011). Making things hard on yourself, but in a good way: creating desirable
difficulties to enhance learning. In M. A. Gernsbacher, R. W. Pew, L. M. Hough, & J. R. Pomerantz (Eds.),
Psychology and the real world: essays illustrating fundamental contributions to society (pp. 5664). New
York: Worth Publishers.
Bjork, R. A., Dunlosky, J., & Kornell, N. (2013). Self-regulated learning: beliefs, techniques, and illusions.
Annual Review of Psychology, 64, 417444. doi:10.1146/annurev-psych-113011-143823.
Brown, A. L., & Smiley, S. S. (1978). The development of strategies for studying texts. Child Development, 49,
Callender, A. A., & McDaniel, M. A. (2009). The limited benefits of rereading educational texts. Contemporary
Educational Psychology, 34,3041.
Craik, F. I. M., & Lockhart, R. S. (1972). Levels of processing: a framework for memory research. Journal of
Verbal Learning & Verbal Behavior, 11,671684.
Dempster, F. N. (1996). Distributing and managing the conditions of encoding and practice. In E. L.Bjork and R.
A. Bjork (Eds.), Handbook of perception and cognition. Vol. 10, Memory, pp. 317-344. New York:
Academic Press.
Dunlosky, J., Rawson, K. A., Marsh, E. J., Nathan, M. J., & Willingham, D. T. (2013). Improving students
learning and comprehension: promising directions from cognitive and educational psychology.
Psychological Science in the Public Interest, 14,458. doi:10.1177/1529100612453266.
Fass, W., & Schumacher, G. M. (1978). Effects of motivation, subject activity, and readability on the retention of
prose materials. Journal of Educational Psychology, 70,803807.
Fowler, R. L., & Barker, A. S. (1974). Effectiveness of highlighting for retention of text material. Journal of
Applied Psychology, 59, 358364.
Hintzman, D. L. (1974). Theoretical implications of the spacing effect. In R. L. Solso (Ed.), Theories in cognitive
psychology: the Loyola Symposium (pp. 7799). Postomac, MD: Erlbaum.
Hoon, P. W. (1974). Efficacy of three common study methods. Psychological Reports, 35, 10571058.
Educ Psychol Rev (2015) 27:6978 77
Author's personal copy
Howe, M. J., & Singer, L. (1975). Presentation variables and studentsactivities in meaningful learning. British
Journal of Educational Psychology, 45, 5261.
Idstein, P., & Jenkins, J. R. (1972). Underlining versus repetitive reading. Journal of Educational Research, 65,
Johnson, L. L. (1988). Effects of underlining textbook sentences on passage and sentence retention. Reading
Research and Instruction, 28,1832.
Karpicke, J. D., Butler, A. C., & Roediger, H. L. (2009). Metacognitive strategies in student learning: do students
practice retrieval when they study on their own? Memory, 138,471479.
Koriat, A., & Bjork, R. A. (2005). Illusions of competence in monitoring ones knowledge during study. Journal
of Experimental Psychology: Learning, Memory, and Cognition, 31, 187194.
Kornell, N. (2009). Optimising learning using flashcards: spacing is more effective than cramming. Applied
Cognitive Psychology, 23, 12971317. doi:10.1002/acp.1537.
Kornell, N., & Bjork, R. A. (2007). The promise and perils of self-regulated study. Psychonomic Bulletin &
Review, 14, 219224.
Kornell, N., & Bjork, R. A. (2009). A stability bias in human memory: overestimating remembering and
underestimating learning. Journal of Experimental Psychology: General, 138,449468.
Leutner, D., Leopold, C., & Den Elzen-Rump, V. (2007). Self-regulated learning with a text-highlighting
strategy: a training experiment. Zeitschrift Für Psychologie/Journal of Psychology, 215(3), 174182. doi:
Nist, S. L., & Hogrebe,M. C. (1987). The role of underlining and annotating inremembering textual information.
Reading Research and Instruction, 27, 1225.
Nist, S. L., & Simpson, M. L. (1988). The effectiveness and efficiency of training college students to annotate
and underline text. National Reading Conference Yearbook, 37,251257.
Peterson, S. E. (1992). The cognitive functions of underlining as a study technique. Reading Research and
Instruction, 31, 4956.
Rickards, J., & August, G. (1975). Generative underlining strategies in prose recall. Journal of Educational
Psychology, 67,860865.
Rickards, J., & Denner, P. R. (1979). Depressive effects of underlining and adjunct questions on childrens recall
of text. Instructional Science, 8,8190.
Roediger, H. L., & Karpicke, J. D. (2006). The power of testing memory: basic research and implications for
educational practice. Perspectives on Psychological Science, 1,181210. doi:10.1111/j.1745-6916.2006.
Silvers, V. L., & Kreiner, D. S. (1997). The effects of pre-existing inappropriate highlighting on reading
comprehension. Reading Research and Instruction, 36,217223.
Smith, S. M., Glenberg, A., & Bjork, R. A. (1978). Environmental context and human memory. Memory &
Cognition, 6, 342353.
Sobel, H. S., Cepeda, N. J., & Kapler, I. V. (2011). Spacing effects in real-world classroom vocabulary learning.
Applied Cognitive Psychology, 25,763767. doi:10.1002/acp.1747.
Stordahl, K. E., & Christensen, C. M. (1956). The effect of study techniques on comprehension and retention.
Journal of Educational Research, 49,561570.
Wade, S. E., & Trathen, W. (1989). Effect of self-selected study methods on learning. Journal of Educational
Psychology, 81,4047.
Wallace, W. P. (1965). Review of the historical, empirical, and theoretical status of the von Restorff phenomenon.
Psychological Bulletin, 63, 410424.
78 Educ Psychol Rev (2015) 27:6978
Author's personal copy
... The use of graphic organizers, for instance, has proven to be an effective instructional strategy to improve reading comprehension of students with ld (Kim et al., 2004). Moreover, explicit instruction on how to highlight texts benefits students' strategic reading performance (Heyne et al., 2020;Leutner et al., 2007;Yue et al., 2015). ...
... Highlighting texts is a cognitive artifact of students' strategic reading performance (Caverly et al., 2000;Heyne et al., 2020;Kobayashi, 2005;Leutner et al., 2007). Thus, highlighting texts represents students' perception of important and relevant ideas (Yue et al., 2015). In the context of this study, these highlighted ideas were related to historical thinking. ...
Full-text available
Document-based history instruction (DBI) was implemented in a middle school special education setting to promote the development of disciplinary cognitive processing and higher order thinking using historical thinking as a framework for students with learning differences (ld). A convergent mixed methods action research design was utilized to explore a) how DBI influenced students’ disciplinary cognitive processing and higherorder thinking when reading multiple historical documents b) the affordances and constraints of using DBI in a special education classroom. Using quantitative data sources (e.g., highlighted documents, reading comprehension responses, and an ondemand writing task) and additional qualitative data (e.g., interviews, fieldnotes, classroom artifacts, and audio recordings), we uncovered insightful evidence to support the potential of DBI for students with ld. In one academic year, when reading multiple historical sources, students learned how to efficiently highlight content knowledge and academic vocabulary, interrogate historical sources, and corroborate information from different sources.
... Since the seminal work of Fowler and Barker (1974), which provided first indications that active text-highlighting supported text comprehension, two distinct mechanisms have been discussed as to why (see e.g. Dunlosky et al. 2013;Winchell, Lan, and Mozer 2020): First, active text-highlighting might serve an encoding function, because readers have to make active decisions about which parts of the text to highlight (Ponce and Mayer 2014;Yue et al. 2014). Second, highlighted text might serve a retrieval function, because highlighted text stands out (or 'pops out'; e.g. ...
This study aimed to investigate the roles of a text-highlighting tool and readers' re-reading behaviour in their integrated understanding of multiple documents. University students (N = 95) read five partly conflicting documents on a health-related issue on a touch display with or without a text-highlighting tool. Integrated understanding of documents was assessed by the number of intertextual connections in essays written after reading and by a source-content mapping task. The provision of the text-highlighting tool resulted in longer initial reading times even when subtracting the time taken for highlighting, but shorter re-reading times, particularly for participants with a high number of re-readings. Further, only for participants with a high number of re-readings, the provision of the text-highlighting tool resulted in more intertextual connections than when no text-highlighting tool was provided. Participants' source-content integration was positively related to the number of re-readings, regardless of whether the text-highlighting tool was provided. Finally, additional exploratory eye-tracking analyses revealed that for two out of the five documents, participants in the with-highlighting condition focused on significantly smaller parts of the documents during re-reading than controls.
... A metacognitive approach in education encourages participants to communicate and engage in directly relevant goals, including abilities, attitudes, self-belief, principles and understandings (Phelps et al., 2004). Research has explored metacognitive beliefs of students in areas such as self-confidence (Kleitman & Gibson, 2011), study strategies (Yue et al., 2015), and student mental health (Valizade et al., 2013). Specifically, in relation to ICT in education, much of the literature relates to the contribution of metacognitive strategy and skill development to student learning (Cadamuro et al., 2019;Maccann-Alfaro et al., 2019;Wu & Peng, 2017). ...
... All pairs were presented using the same font size with the only perceptual difference being the difference in highlight presentation. We selected this manipulation, as under some conditions, the use of highlighting can be beneficial to comprehension and learning, as highlighting makes text distinguishable from non-highlighted material (Fowler & Barker, 1974;Yue et al., 2015). ...
Full-text available
Judgments of learning (JOL) are often used to assess memory monitoring at encoding. Participants study a cue-target word pair (e.g., mouse-cheese) and are asked to rate the probability of correctly recalling the target in the presence of the cue at test (e.g., mouse -?). Prior research has shown that JOL accuracy is sensitive to perceptual cues. These cues can produce metamemory illusions in which JOLs overestimate memory, such as the font-size effect (Rhodes & Castel, 2008), which occurs when participants inflate JOLs for pairs presented in large versus small fonts without a concomitant increase to recall. The present study further tests the font-size effect and examines whether other perceptual manipulations can affect the correspondence between JOLs and recall. Experiments 1A and 1B were designed to replicate the font-size effect and test whether the effect extended to highlighted pairs that were related or unrelated in the same study list. Experiment 2A and 2B examined font size and highlighting effects on JOLs using only unrelated pairs. Finally, Experiment 3 tested whether Sans Forgetica—a perceptually disfluent font designed to improve memory—would result in inflated JOLs and/or recall. Large fonts similarly increased both JOLs and recall relative to small fonts, highlights had no effect on JOLs or recall, and Sans Forgetica font yielded a memory cost (though no effect on JOLs). Collectively, perceptually fluent and disfluent study pairs do not appear to inflate JOLs relative to subsequent recall.
... This possibility cannot be ruled out and future research could use a between-subjects design where one group of participants performed text highlighting and another group read the text and answered direct attitude questions. Certainly, the possibility of testing bias does exist considering the advantages of highlighting in learning, where the act of deciding what to mark and what not to mark may lead students to process textual information at a deeper and more evaluative level than they would when simply reading it (Yue, Storm, Kornell, & Bjork, 2015). Drawing on the Elaboration Likelihood Model of attitude change (Petty & Cacioppo, 1986a), this could imply that text highlighting increases the probability that participants will follow the central route to persuasion, which compared with the peripheral route is associated with attitude change that is more enduring and predictive of behaviour (Petty & Cacioppo, 1986b). ...
Attitude measurement occupies a central position in consumer research. Concerns over the validity and reliability of traditional measures have motivated the development of alternative approaches. The present research introduces text highlighting as a method for measurement of explicit attitudes using a case study on vertical farming (VF) with 837 UK consumers. They participated in an online survey, where they read a text about VF and used highlighting functions to mark text as ‘like’ and ‘dislike.’ Consumers approached the task in a systematic and logical way and desirable aspects of VF were frequently highlighted as ‘like’, whereas undesirable aspects were more frequently highlighted as ‘dislike’. The text highlighting responses were summarised using word clouds, frequency tables and through sentiment scores to reveal an overall positive attitude to VF among participants. Sentiment scores enabled the identification of consumer segments with interpretable differences in their attitude towards VF. Two approaches to method validation – comparison with direct attitude questions and consumer profiling – further confirmed the potential of the text highlighting method. The sentiment of specific sentences in the text highlighting task matched results from self-reported attitudinal based on Likert scales. Consumer segments with different sentiment in the text highlighting task also differed in their food technology neophobia scores in the expected direction. Future research should investigate methodological aspects of text highlighting and explore its suitability to other applications.
... The second strategy involved using sticky notes to create a place on the page for students to jot down reactions, questions, and connections made as they read each article (Daniels & Zemelman, 2014). Highlighters were also provided for students who preferred to mark text this way (Yue, Storm, Kornell, & Bjork, 2015). ...
Full-text available
Part Two of this lesson further develops students' literacy skills in the areas of critical reading and writing. Students apply their scientific knowledge and basic argumentation skills to determine the safety of a single traffic intersection. Students learn how to critically read scholarly journal articles and construct a formal written argument. Students will broaden the scope of their argument from a single intersection to the best practices for enforcement of traffic laws. Moving past emotionally-charged opinions, students use scientific evidence to inform and establish an evidence-based position on the use of cameras to impact the behavior of motorists.
ive summarization aims to generate a concise summary covering salient content from single or multiple text documents. Many recent abstractive summarization methods are built on the transformer model to capture long-range dependencies in the input text and achieve parallelization. In the transformer encoder, calculating attention weights is a crucial step for encoding input documents. Input documents usually contain some key phrases conveying salient information, and it is important to encode these phrases completely. However, existing transformer-based summarization works did not consider key phrases in input when determining attention weights. Consequently, some of the tokens within key phrases only receive small attention weights, which is not conducive to encoding the semantic information of input documents. In this paper, we introduce some prior knowledge of key phrases into the transformer-based summarization model and guide the model to encode key phrases. For the contextual representation of each token in the key phrase, we assume the tokens within the same key phrase make larger contributions compared with other tokens in the input sequence. Based on this assumption, we propose the Key Phrase Aware Transformer (KPAT), a model with the highlighting mechanism in the encoder to assign greater attention weights for tokens within key phrases. Specifically, we first extract key phrases from the input document and score the phrases’ importance. Then we build the block diagonal highlighting matrix to indicate these phrases’ importance scores and positions. To combine self-attention weights with key phrases’ importance scores, we design two structures of highlighting attention for each head and the multi-head highlighting attention. Experimental results on two datasets (Multi-News and PubMed) from different summarization tasks and domains show that our KPAT model significantly outperforms advanced summarization baselines. We conduct more experiments to analyze the impact of each part of our model on the summarization performance and verify the effectiveness of our proposed highlighting mechanism.
Full-text available
The present study examines the existing published research about the effectiveness of learner-generated highlighting and instructor-provided highlighting on learning from text. A meta-analysis was conducted of scientifically rigorous experiments comparing the learning outcomes (i.e., performance on memory and/or comprehension tests) of students (i.e., college students and/or K-12 students) who read an academic text with or without being asked to highlight important material (i.e., with or without learner-generated highlighting) or who read an academic text with or without the important material already being highlighted (i.e., with or without instructor-provided highlighting). We found 36 published articles that met these criteria ranging from the years 1938 to 2019, which generated 85 effect sizes. The results showed that learner-generated highlighting improved memory but not comprehension, with average effect sizes of 0.36 and 0.20, respectively; and instructor-provided highlighting improved both memory and comprehension, both with an average effect size of 0.44. Learner-generated highlighting improved learning for college students but not for school students, with average effect sizes of 0.39 and 0.24, respectively; and instructor-provided highlighting improved learning for both college and school students, with average effect sizes of 0.41 and 0.48, respectively. We discuss the theoretical and practical implications of these findings.
Adaptive learning solutions generally assume curricular, instructional, content, and/or learning models to be known in advance. This paper describes an ongoing research project being conducted by the Center for Inclusive Software for Learning (CISL), funded by the U.S. Department of Education, that is developing a suite of open-source tools designed for diverse learners by making digital educational materials—including open educational resources (OER)—accessible, flexible, and engaging. The Clusive Reader at the heart of CISL leverages Universal Design for Learning (UDL) principles and includes a number of adaptive features designed to operate without prior knowledge of how the materials support learning goals. These adaptive features include content complexity leveling, inclusion of summaries and highlighted main ideas, dictionary and glossary tools, and just-in-time tips and suggestions regarding preference setting and use of learning tools. Information to support adaptivity heuristics comes from user interactions, an onboarding skills discovery process, and embedded affective, comprehension, and learning goal prompts and game-like vocabulary challenges. Per UDL, adaptivity in Clusive is designed to support not only immediate learning goals, but also the building of “expert learners” who independently choose the most effective content, contexts, and supports. As such, adaptivity has been designed as a learning scaffold, with the intention of gradual reduction over time, in two ways. First, adaptivity operates transparently, providing students with insights as to why choices and recommendations are made. Second, students are allowed to override system decisions to promote self-agency. Evaluation of student impressions and efficacy are being conducted as part of the CISL project. This paper provides the research foundations and philosophical approach taken for this effort, describes the current adaptive learning features being designed, and concludes with remarks on next steps and challenges being faced.
Full-text available
Integrating multiple representations into a coherent mental model is one of the challenges when learning with multimedia. In this experimental study (N=173), we examined how highlighting corresponding information in text-graph learning material can help higher education students to make the necessary connections and improve learning outcomes in two domains. We compared a control condition to a signaling condition with given highlights and an active signaling condition where students were asked to visually highlight corresponding text and graph information themselves. There was no overall benefit of given signals or active signaling. We discuss prior knowledge and the quality of learner-generated signals as possible explanations. For biology learning material , prior knowledge moderated the effect of active signaling. In economics, the effect of prior knowledge was mediated by the quality of learner-generated correspondences. Our findings suggest that different methods of supporting text-graph integration might be useful for different students and learning material. This article is protected by copyright. All rights reserved.
Full-text available
Many students are being left behind by an educational system that some people believe is in crisis. Improving educational outcomes will require efforts on many fronts, but a central premise of this monograph is that one part of a solution involves helping students to better regulate their learning through the use of effective learning techniques. Fortunately, cognitive and educational psychologists have been developing and evaluating easy-to-use learning techniques that could help students achieve their learning goals. In this monograph, we discuss 10 learning techniques in detail and offer recommendations about their relative utility. We selected techniques that were expected to be relatively easy to use and hence could be adopted by many students. Also, some techniques (e.g., highlighting and rereading) were selected because students report relying heavily on them, which makes it especially important to examine how well they work. The techniques include elaborative interrogation, self-explanation, summarization, highlighting (or underlining), the keyword mnemonic, imagery use for text learning, rereading, practice testing, distributed practice, and interleaved practice. To offer recommendations about the relative utility of these techniques, we evaluated whether their benefits generalize across four categories of variables: learning conditions, student characteristics, materials, and criterion tasks. Learning conditions include aspects of the learning environment in which the technique is implemented, such as whether a student studies alone or with a group. Student characteristics include variables such as age, ability, and level of prior knowledge. Materials vary from simple concepts to mathematical problems to complicated science texts. Criterion tasks include different outcome measures that are relevant to student achievement, such as those tapping memory, problem solving, and comprehension. We attempted to provide thorough reviews for each technique, so this monograph is rather lengthy. However, we also wrote the monograph in a modular fashion, so it is easy to use. In particular, each review is divided into the following sections: General description of the technique and why it should work How general are the effects of this technique? 2a. Learning conditions 2b. Student characteristics 2c. Materials 2d. Criterion tasks Effects in representative educational contexts Issues for implementation Overall assessment The review for each technique can be read independently of the others, and particular variables of interest can be easily compared across techniques. To foreshadow our final recommendations, the techniques vary widely with respect to their generalizability and promise for improving student learning. Practice testing and distributed practice received high utility assessments because they benefit learners of different ages and abilities and have been shown to boost students’ performance across many criterion tasks and even in educational contexts. Elaborative interrogation, self-explanation, and interleaved practice received moderate utility assessments. The benefits of these techniques do generalize across some variables, yet despite their promise, they fell short of a high utility assessment because the evidence for their efficacy is limited. For instance, elaborative interrogation and self-explanation have not been adequately evaluated in educational contexts, and the benefits of interleaving have just begun to be systematically explored, so the ultimate effectiveness of these techniques is currently unknown. Nevertheless, the techniques that received moderate-utility ratings show enough promise for us to recommend their use in appropriate situations, which we describe in detail within the review of each technique. Five techniques received a low utility assessment: summarization, highlighting, the keyword mnemonic, imagery use for text learning, and rereading. These techniques were rated as low utility for numerous reasons. Summarization and imagery use for text learning have been shown to help some students on some criterion tasks, yet the conditions under which these techniques produce benefits are limited, and much research is still needed to fully explore their overall effectiveness. The keyword mnemonic is difficult to implement in some contexts, and it appears to benefit students for a limited number of materials and for short retention intervals. Most students report rereading and highlighting, yet these techniques do not consistently boost students’ performance, so other techniques should be used in their place (e.g., practice testing instead of rereading). Our hope is that this monograph will foster improvements in student learning, not only by showcasing which learning techniques are likely to have the most generalizable effects but also by encouraging researchers to continue investigating the most promising techniques. Accordingly, in our closing remarks, we discuss some issues for how these techniques could be implemented by teachers and students, and we highlight directions for future research.
Full-text available
A powerful way of improving one's memory for material is to be tested on that material. Tests enhance later retention more than additional study of the material, even when tests are given without feedback. This surpris- ing phenomenon is called the testing effect, and although it has been studied by cognitive psychologists sporadically over the years, today there is a renewed effort to learn why testing is effective and to apply testing in educational settings. In this article, we selectively review laboratory studies that reveal the power of testing in improving re- tention and then turn to studies that demonstrate the basic effects in educational settings. We also consider the related concepts of dynamic testing and formative assess- ment as other means of using tests to improve learning. Finally, we consider some negative consequences of testing that may occur in certain circumstances, though these negative effects are often small and do not cancel out the large positive effects of testing. Frequent testing in the classroom may boost educational achievement at all levels of education. In contemporary educational circles, the concept of testing has a dubious reputation, and many educators believe that testing is overemphasized in today's schools. By ''testing,'' most com- mentators mean using standardized tests to assess students. During the 20th century, the educational testing movement produced numerous assessment devices used throughout edu- cation systems in most countries, from prekindergarten through graduate school. However, in this review, we discuss primarily the kind of testing that occurs in classrooms or that students engage in while studying (self-testing). Some educators argue
30 college males in liberal arts curricula were assigned to study passages by 3 different methods: reading, reading with underlining, and reading with note taking. When verbal intelligence (Scholastic Aptitude Test-Verbal) was held constant, the 3 groups did not differ on questions about comprehension of passages. The findings question the assumed value of underlining and note-taking during study and imply that future research might better be aimed at improving verbal rather than study skills in undergraduate populations.
Two experiments were performed to determine whether underlining was superior to repetitive reading on a completion test following a delayed review period. In the first study there were four conditions making up a 2 × 2 analysis of variance. The conditions were underlining vs repetitive reading × 9-minute vs 4½-minute review. In the second study there were two conditions, underlining ca repetitive reading. In the first study there was no significant difference for underlining vs repetitive reading, however, the difference between the 9- and 4½-minute review was significant (p <.05). In the second study there were no significant differences.
Below average college freshmen read an 1844‐word selection about the Kalahari Desert. The experimental group underlined one sentence per paragraph, and the control group only read the passage. Afterwards subjects took a multiple‐choice test of items drawn from superordinate and subordinate sentences. While overall passage retention was not improved, underlined sentences were recalled better than nonunderlined sentences, and superordinate sentences, whether or not underlined, were retained better than subordinate sentences by the experimental but not by the control group. When a review was added in a second experiment, passage retention was still not improved. This time both experimental and control groups retained superordinate sentences better than subordinate sentences. Underlining subordinate sentences improved their retention even above that achieved by the control group. Results suggest that without review, underlining helped these below average students sort out superordinate ideas. With review, underlining subordinate sentences helped their retention without decreasing retention of superordinate sentences.
Forty-five university students participated in a computer-based training program on self-regulated learning from expository text. The training program introduced students to a learning strategy helping them identify and highlight important text information. Students were randomly assigned to three treatment groups: (1) no training at all, (2) training in highlighting only, or (3) combined training in both highlighting and self-regulation. After completing the training, students were instructed to read an instructional text and apply the trained strategies. The extent to which they applied the strategies while reading the text was assessed, and the amount of knowledge and comprehension they had acquired and recalled from the text was measured. Results show that students in the combined training condition outperformed their counterparts in the learning strategy training condition, who in turn outperformed those with no training at all. The results are in line with recent self-regulated learning theories, which state that, in addition to teaching students specific cognitive learning strategies, it is worth training them to monitor and regulate their strategy use.
In a series of 3 experiments, the strategies of children and college students were examined as they attempted to study texts. College students, under various intentional learning instructions, displayed a clear diagnostic pattern. Following extended study they improved recall of important, but not unimportant, elements of texts. Eleventh and twelfth graders conformed to the adult pattern, but fifth through eighth graders were not as efficient. Older students benefited from increased study time because they possessed the necessary knowledge concerning the importance of text segments to enable them to concentrate on the essential. Younger students, not so prescient, did not concentrate exclusively on the important units, for they did not know what they were. Age was not the sole determinant of performance, for some students at each age spontaneously adopted the strategies of underlining or note-taking. Those who did concentrated on the important elements and subsequently approached the adult-like pattern in recall; those who did not displayed the immature pattern, even if induced to adopt 1 of the strategies.