Content uploaded by Karina Agadzhanyan
Author content
All content in this area was uploaded by Karina Agadzhanyan on Dec 03, 2021
Content may be subject to copyright.
RESEARCH ARTICLE
Learning in double time: The effect of lecture video speed on
immediate and delayed comprehension
Dillon H. Murphy | Kara M. Hoover | Karina Agadzhanyan |
Jesse C. Kuehn | Alan D. Castel
Department of Psychology, University of
California, Los Angeles, California, USA
Correspondence
Dillon H. Murphy, Department of Psychology,
University of California, Los Angeles, Los
Angeles, CA 90095, USA.
Email: dmurphy8@ucla.edu
Abstract
We presented participants with lecture videos at different speeds and tested immedi-
ate and delayed (1 week) comprehension. Results revealed minimal costs incurred by
increasing video speed from 1x to 1.5x, or 2x speed, but performance declined
beyond 2x speed. We also compared learning outcomes after watching videos once
at 1x or twice at 2x speed. There was not an advantage to watching twice at 2x
speed but if participants watched the video again at 2x speed immediately before the
test, compared with watching once at 1x a week before the test, comprehension
improved. Thus, increasing the speed of videos (up to 2x) may be an efficient strat-
egy, especially if students use the time saved for additional studying or rewatching
the videos, but learners should do this additional studying shortly before an exam.
However, these trends may differ for videos with different speech rates, complexity
or difficulty, and audiovisual overlap.
KEYWORDS
comprehension, metacognition, online learning, video speed
1|INTRODUCTION
The use of technology in the classroom and as a learning aid has
become ubiquitous, particularly during the COVID-19 pandemic
(Pokhrel & Chhetri, 2021; see also Belt & Lowenthal, 2021). Addition-
ally, even before the COVID-19 pandemic, many teachers and profes-
sors turned to asynchronous online classes to disseminate course
material (Barbour, 2013), with lecture videos being the foundation for
learning the material in these classes. Specifically, compared with
quizzes, assignments, discussions, and other course activities, students
spend the most time watching lecture videos (Breslow et al., 2013).
Thus, watching and remembering information from online lecture
videos is imperative for course performance and successful learning.
Although asynchronous online classes provide students with
the flexibility to choose when and how to learn the material,
self-regulating studying can be problematic for some students (see
Boekaerts, 1997; Panadero, 2017; Thiede & De Bruin, 2017; Wong
et al., 2019; see also Bjork et al., 2013). Specifically, without clearly
structured in-person classes, students may struggle to effectively
and efficiently allocate their study time and study choices (see
Zimmerman, 1989, 1990). For example, students who have less time
to devote to academia (e.g., working a part-time job, family obliga-
tions) may not allocate sufficient study time to their coursework, lead-
ing to impaired memory for to-be-learned material.
In addition to other obligations, students' study regulation may be
influenced by instruction pace, perception of the difficulty of the
material, and motivation to learn (Sinha et al., 2014). In light of these
influences to study regulation, asynchronous online lectures may
allow students to bolster their studying efficacy by allowing them to
customize when and how they watch pre-recorded lectures. Com-
pared with live lectures, students sometimes claim that their needs
are better satisfied when lectures are pre-recorded such that they can
learn and retain more of the information, find more time for other
The experiments reported in this article were formally preregistered and the stimuli, data, and
analysis code have been made available on the Open Science Framework here.
Received: 26 April 2021 Revised: 7 October 2021 Accepted: 10 November 2021
DOI: 10.1002/acp.3899
Appl Cognit Psychol. 2021;1–14. wileyonlinelibrary.com/journal/acp © 2021 John Wiley & Sons Ltd. 1
activities, manage stress, and stay focused (Cardall et al., 2008). Spe-
cifically, one way that students can conserve their time and cope with
the demands of online courses is by watching lectures at an increased
playback speed. By watching lectures at a faster speed, students can
study the same amount of material in a condensed amount of time,
giving them more time to allocate to other courses and activities.
Watching asynchronous lectures at a higher speed may be a use-
ful and efficient study strategy if it results in similar or better compre-
hension than when watching lectures at a normal speed; however,
there has been some disagreement regarding the effect of video
speed on comprehension. For example, some work has found that
increasing the speed of videos can preserve or even enhance compre-
hension (Lang et al., 2020; Nagahama & Morita, 2017; Wilson
et al., 2018) while others suggest that increased speed impairs com-
prehension (Foulke & Sticht, 1969; Song et al., 2018; Vemuri
et al., 2004). These disagreements may be the product of using limited
stimuli, measuring comprehension immediately after watching the
video (rather than a delayed test), allowing for note-taking and partici-
pant control of the videos (i.e., pausing, rewinding), using very short
video clips (i.e., 20 s), and small sample sizes.
Regardless of the veridical effect of video speed on comprehen-
sion, students may believe that watching videos at an increased speed
does not impair learning or may even be an advantageous study
technique (see Wilson et al., 2018). Thus, examining learners' aware-
ness of their memory processes (i.e., metacognition; Nelson &
Narens, 1990; see also Dunlosky et al., 2016; Nelson, 1996) when
watching asynchronous lectures is crucial in understanding how stu-
dents monitor and regulate their learning. Specifically, students'
potentially misguided beliefs about the effect of video speed on learn-
ing could result in less effective regulation of study time and poor
learning outcomes.
If students watch lecture videos at a faster than normal speed, this
could reduce subsequent memory for the material as a result of a cognitive
overload (i.e., the amount of informationthatcanbeheldinworkingmem-
ory at a given time). According to the Cognitive Load Theory (see
Sweller, 1988, 1989), when new information is being learned, this informa-
tion is stored in working memory before being transferred to long-term
memory. Although long-term memory capacity is generally considered to
be relatively limitless, working memory capacity is much more restricted
(Baddeley & Hitch, 1974; Cowan, 2010; Gilchrist et al., 2008; Miller, 1956;
also see the Cognitive Theory of Multimedia Learning; Mayer, 2002). Thus,
a surplus of information in working memory can hinder transfer to long-
term memory, preventing learning (Sweller et al., 2011). Specifically, both
the difficulty or complexity of the information (intrinsic cognitive load) and
how the information is presented (extraneous cognitive load) can increase
cognitive load and impair learning (Paas et al., 2003, 2004; Sweller
et al., 1998; van Merriënboer & Sweller, 2005). Applied to playback speed,
if lecture videos are watched at a rate that overwhelms our limited cogni-
tive resources, learners' comprehension of the material may be impaired.
Additionally, different modalities of instruction (i.e., audio-visual)
may be differentially impacted by increased playback speeds. Specifi-
cally, the transient information effect suggests that complex informa-
tion should not be provided in auditory form as working memory is
likely to be overwhelmed. Rather, if to-be-learned material is particu-
larly complex, it may be better learned via reading (whereby critical
information can be re-accessed) rather than listening (see Leahy &
Sweller, 2011). Thus, as to-be-learned information becomes increas-
ingly complex, the potential costs of increased playback speed may be
more pronounced.
At normal speed, lecture videos are easily comprehensible but
increasing the playback speed increases the number of words spoken
per minute, potentially making the videos too cognitively taxing. For
example, humans generally speak at a rate of 150 words per minute
(Peelle & Davis, 2012), and prior work suggests that speech compre-
hension begins to decline at around 275 words per minute if the infor-
mation is encoded just audibly (see Foulke & Sticht, 1969). However,
audiovisual materials (i.e., videos that consist of both visual and audi-
ble content) may be more comprehensible at increased presentation
speeds due to benefits from the visually presented information. Thus,
high-speed lecture videos may lead to deficits in later remembering as
a result of decreased speech comprehension and increased cognitive
load, but the visual component of lecture videos could compensate
for these potential deficits (e.g., Pastore & Ritzhaupt, 2015). However,
increased video speeds would result in less time to encode any visu-
ally presented information. Taken together, if lecture videos are
played at too fast of a speed, memory for the material may be
impaired.
2|THE CURRENT STUDY
In the current study, we investigated how watching lecture videos at
various speeds affects comprehension and metacognitive monitoring
of learning. Specifically, in Experiment 1, participants watched lecture
videos at either normal (1x) speed or increased speeds (1.5x, 2x, or
2.5x) and were tested on the video content both immediately and
after a delay (1 week). To further investigate the most efficient
methods for watching lecture videos, in Experiment 2, we examined
whether watching a video twice at 2x speed results in better learning
outcomes than watching a single time at 1x speed. In Experiment
3, we tested how different study schedules (watching first at normal
speed and again at 2x speed or first at 2x speed and then again at nor-
mal speed) affect comprehension. Ultimately, the results of these
experiments will provide insight into effective and efficient methods
of learning when watching asynchronous online lecture videos.
3|EXPERIMENT 1
In Experiment 1, we investigated how memory for information from
asynchronous lecture videos is affected by playback speed. Specifi-
cally, participants watched lecture videos on real estate appraisals and
the history of the Roman Empire at either 1x speed or increased
speeds (1.5x, 2x, or 2.5x) and were tested on the video content both
immediately after watching and after a delay (1 week). Additionally,
we solicited metacognitive predictions of immediate and delayed
2MURPHY ET AL.
performance to determine whether participants were metacognitively
aware of any potential effects of video speed on comprehension. Although
we expected that immediate comprehension may be preserved or even
enhanced at faster video speeds (see Lang et al., 2020; Nagahama &
Morita, 2017), after a delay, we expected increased video speeds to lead
to poorer memory performance compared with normal speed. In terms of
participants' metacognitive predictions of performance, consistent with a
stability bias in memory (cf. Kornell et al., 2011), we expected participants
to predict that both immediate and delayed retention would be minimally
affected by video speed, as some work suggests that participants may not
use retention interval information when making metacognitive judgments
(cf. Koriat et al., 2004; Kornell et al., 2011).
3.1 |Method
3.1.1 | Participants
After exclusions, participants were 231 undergraduate students (aged
18–41: M=20.83, SD =2.80) recruited from the University of Califor-
nia, Los Angeles (UCLA) Human Subjects Pool. All participants were flu-
ent in English and 68% were native English speakers. Participants were
tested online and received course credit for their participation. Partici-
pants were excluded from analysis if they admitted to cheating
(e.g., looking up answers) in a post-task questionnaire (participants were
told they would still receive credit if they cheated). This exclusion pro-
cess resulted in two exclusions. Additionally, at the end of the study, if
participants self-reported having prior expertise on either video topic
(four participants reported having expertise on appraisals and 19 reported
having expertise on the Roman Empire), their scores on that topic were
excluded from analysis. An a priori power analysis indicates that for an
omnibus, one-way analysis of variance (ANOVA) with four groups (video
speed), assuming alpha =.05, power =.80, 212 participants would be
needed to reliably detect a small effect (η2=.05).
3.1.2 | Materials
Participants watched two lecture videos judged by the experimenters
to likely present novel material to participants. The videos were on
real estate appraisals (12 min and 56 s with 2031 spoken words) and
the history of the Roman Empire (14 min and 27 s with 2403 spoken
words). The videos were accessed through YouTube and modified to
play at four different speeds (1x, 1.5x, 2x, and 2.5x; see Table 1 for
video durations and speech rates at each speed). Each video consisted
of presentation slides along with a video of the lecturer on the left
side of the screen; the videos did not contain captions or subtitles.
To measure learning, we created two comprehension tests
(20 questions each; one for immediate comprehension and one for
delayed comprehension) for each lecture consisting of multiple-choice
and true or false questions; the multiple-choice questions contained
four options for participants to select from. Of the 80 total compre-
hension questions, 51 of the answers appeared verbally only, 1 only
appeared visually, and 28 appeared both verbally and visually. Test
order was counterbalanced between the immediate and delayed ses-
sions. Comprehension was calculated as the proportion of questions
answered correctly across video topics.
3.1.3 | Procedure
The procedure used in Experiment 1 is shown in Figure 1. Participants
were randomly assigned to watch both videos at either 1x speed
(n=57), 1.5x speed (n=58), 2x speed (n=59), or 2.5x speed
(n=57). Participants were told that they would be watching a short
video and then taking a comprehension test on the material covered
in the video. They were also instructed to watch the video in full-
screen mode and not to pause the video or take any notes. Partici-
pants then watched the video on real estate appraisals, made a predic-
tion of their immediate test performance, took a comprehension test,
and made a prediction of their performance on a similar exam in
1 week. When making predictions, participants were asked how many
of the 20 questions they expected to get correct.
Next, participants repeated this procedure for the Roman Empire
lecture at the same speed as the first video. After a 1-week delay, par-
ticipants were given 1 week to complete the second part of the exper-
iment. In the second (final) part of the experiment, before taking a
similar comprehension test on appraisals, participants predicted their
performance. Participants were then tested on real estate appraisals
and repeated this procedure for the Roman Empire content. A control
group
1
(n=123) who did not watch the lecture videos also completed
the comprehension tests to serve as a comparison group to the exper-
imental groups watching videos at either 1x, 1.5x, 2x, or 2.5x speed.
Informed consent was acquired, and the study was completed in
accordance with UCLA's Institutional Review Board.
3.2 |Results
In each experiment, we collapsed predictions and performance across
topics to control for variance in learning different subjects and ensure
that effects were domain general.
TABLE 1 Video duration (minutes, seconds) and words per
minute for each video topic and speed
Real estate
appraisals
History of the
Roman Empire
1x duration 12:56 14:27
1x words per minute 157 166
1.5x duration 8:40 9:38
1.5x words per minute 210 234
2x duration 6:28 7:14
2x words per minute 314 332
2.5x duration 5:10 5:47
2.5x words per minute 393 416
MURPHY ET AL.3
3.2.1 | Predictions
Descriptive statistics for predictions and performance at each video
speed (1x, 1.5x, 2x, and 2.5x) are shown in Table 2. To investigate pos-
sible differences in participants' predictions of performance, we com-
puted a four (video speed: 1x, 1.5x, 2x, 2.5x) x 3 (time: immediate,
delayed, immediately before the delayed test) mixed ANOVA. Results
revealed a main effect of time (F[2, 404] =134.59, p< .001, η2=.39)
such that participants' predictions of immediate performance were
greater than their predictions of delayed performance, (p
bonf
< .001,
d=.78) as well as participants' predictions of performance immedi-
ately before the delayed test, (p
bonf
< .001, d=1.09); additionally,
participants' predictions of delayed performance were greater than
their predictions of performance immediately before the delayed test
(p
bonf
< .001, d=.26). However, results did not reveal a main effect
of video speed (F[3, 202] =.95, p=.418, η2=.01) but there was an
interaction between time of predictions and video speed (F[6, 404] =
2.50, p=.022, η2=.02). A post-hoc ANOVA indicated that for
the predictions regarding the immediate test, participants in the 2.5x
speed group expected to do worse than participants watching at
normal speed (p
bonf
=.022, d=.19). However, participants in the
2.5x speed group expected to do similarly as other participants on the
delayed test and following the delay, maintained this expectation of
performing similarly as the other groups (all p
bonf
> .609, all d< .08).
3.2.2 | Performance
To examine performance on the immediate and delayed comprehen-
sion tests (see Figure 2), we computed a four (video speed: 1x, 1.5x, 2x,
2.5x) x 2 (time of test: immediate, delayed) mixed ANOVA. Results rev-
ealed a main effect of test time (F[1, 202] =98.96, p<.001,η2=.33)
such that performance on the immediate test (M=.60, SD =.15) was
greater than on the delayed test (M=.54, SD =.15). Additionally,
results revealed a main effect of video speed (F[3, 202] =3.98,
p=.009, η2=.06) such that the 1x group performed better than the
2.5x group (p
bonf
=.004, d=.24) but there were no other pairwise dif-
ferences (all p
bonf
>.385,alld< .13). Moreover, time of test and video
speed did not interact (F[3, 202] =.78, p=.504, η2=.01) such that
learning outcomes for participants watching at various video speeds did
not differ as a function of the time of the test.
3.2.3 | Control group
To further elucidate the effect of video speed on immediate and
delayed comprehension, we collected an additional sample of
123 undergraduates who did not watch the videos. These partici-
pants completed all 80 comprehension questions, and their average
performance (M=.41, SD =.49) is represented by the dashed line in
Figure 2. We also asked these participants at what speed they usu-
ally watch pre-recorded lecture videos and 15% reported watching
at normal speed, 60% reported watching at 1.5x speed, 23%
reported watching at 2x speed, and 3% reported watching at 2.5x
speed (see Figure 3a). Lastly, we asked participants what lecture
video speed they think is the best for learning and 42% selected nor-
mal speed, 49% selected 1.5x speed, 8% selected 2x speed, and 2%
selected 2.5x speed (see Figure 3b).
3.3 |Discussion
In Experiment 1, participants watched lecture videos at either 1x,
1.5x, 2x, or 2.5x speed and took comprehension tests both
FIGURE 1 The sequence of events in each trial in Experiment 1. The first trial used the video on real-estate appraisals and the second trial
used the video on the Roman Empire
TABLE 2 Means and SD (in parentheses) for predictions and performance at each video speed in Experiment 1
Video
speed
Immediate
predictions
Immediate
performance
Delayed
predctions
Delayed
performance
Delayed
immediate
predctions
1x .54 (.19) .65 (.13) .36 (.18) .59 (.13) .31 (.17)
1.5 .49 (.19) .60 (.16) .35 (.20) .54 (14) .29 (.19)
2 .48 (.18) .62 (.14) .32 (.19) .53 (.15) .29 (.12)
2.5x .43 (.20) .55 (.15) .35 (.22) .50 (.15) .32 (.20)
4MURPHY ET AL.
immediately after watching the videos and after a 1-week delay. We
also asked participants to predict their immediate and delayed perfor-
mance to determine whether participants are metacognitively aware
of video speed's potential effects on learning. Surprisingly, results rev-
ealed that video speed had little effect on both immediate and del-
ayed comprehension such that learning was not significantly impaired
in participants watching videos at 1.5x and 2x speed; comprehension
was only impaired in participants watching at 2.5x speed. However,
despite some learning impairments, participants who watched the
videos at 2.5x speed still successfully encoded some of the material.
2
Finally, participants' predictions of performance did not show any
significant differences between the 1x, 1.5x, and 2x speeds; the only
pairwise difference was between the immediate performance predic-
tions of the 1x and 2.5x speed groups. Collectively, the results of
Experiment 1 revealed that watching lecture videos at 1.5x or 2x does
not result in learning impairments, and students could save time and
more efficiently learn by watching pre-recorded lectures at faster
speeds, but they should not exceed 2x speed.
4|EXPERIMENT 2A
In Experiment 1, results revealed no significant differences in compre-
hension between participants watching at 1x, 1.5x, and 2x speed. Given
that the cost of watching lectures at an increased speed can be minimal,
it may be advantageous for students to allocate the time saved from
watching videos at a higher speed toward another class or additional
studying (Cermak et al., 1996). For example, rather than watching lec-
tures a single time at normal speed, students may be able to enhance
learning by watching lectures twice (see Greene, 1989; Hintzman, 1976;
Hintzman & Block, 1971; Raajimakers, 2003 for the memory benefits of
repetition). Although at normal speed this would take twice the time, if
participants watch the videos at 2x speed, they may be able to harness
the benefits of repetition without spending additional time studying. In
Experiment 2a, we investigated whether participants could enhance
learning outcomes without spending additional time studying by
watching lecture videos twice at 2x speed compared with once at a nor-
mal speed. We expected participants to predict higher performance and
demonstrate better learning outcomes after repeated study opportuni-
ties (watching twice at 2x speed compared with once at 1x speed).
4.1 |Method
4.1.1 | Participants
After exclusions, participants were 106 undergraduate students (aged
18–36: M=20.52, SD =2.03) recruited from the UCLA Human Sub-
jects Pool. All participants were fluent in English and 67% were native
English speakers. Participants were tested online and received course
credit for their participation. Participants were excluded from analysis
if they admitted to cheating (e.g., looking up answers) in a post-task
questionnaire (participants were told they would still receive credit if
they cheated). This exclusion process resulted in one exclusion.
FIGURE 2 Performance on the
immediate and delayed comprehension
tests as a function of video speed in
Experiment 1. The dashed line represents
the mean performance of participants
who did not watch the videos. Error bars
reflect the SEM
FIGURE 3 Control group participants' reported speed at which
they usually watch lecture videos (a) and the speed that they think is
the best for learning
MURPHY ET AL.5
Additionally, at the end of the study, if participants self-reported hav-
ing prior expertise on either video topic (one participant reported hav-
ing expertise on appraisals and 20 reported having expertise on the
Roman Empire), participants' scores on that topic were excluded from
analysis. An a priori power analysis indicated that for a two-group test
of independent means, assuming alpha =.05, power =.80, for a two-
tailed test, 90 participants would be needed to reliably detect a
medium effect size (d=.60).
4.1.2 | Materials and procedure
The materials were similar to Experiment 1. Participants were ran-
domly assigned to either watch both videos once at 1x speed (n=53)
or twice at 2x speed (n=53). After watching the video on real estate
appraisals either once at 1x speed or twice at 2x speed (participants
watching the videos twice watched them in immediate succession but
were not told that they would watch each video twice), participants
predicted their performance on the comprehension test. Participants
then completed the comprehension test (20 questions) and repeated
this procedure for the Roman Empire video.
4.2 |Results
Predictions of immediate performance and comprehension test perfor-
mance as a function of viewing schedule are shown in Figure 4. To inves-
tigate possible differences in participants' predictions of performance, we
computed an independent samples ttest. Results revealed that partici-
pants watching the videos a single time at 1x speed expected to perform
better on the comprehension test (M=.61, SD =.20) than participants
watching the videos twice at 2x speed (M=.51, SD =.18), (t
[104] =2.50, p=.014, d=.49). However, an independent samples
ttest on test performance did not reveal differences in comprehension
between participants watching the videos a single time at 1x speed
(M=.64, SD =.18) and participants watching the videos twice at 2x
speed (M=.63, SD =.17), (t[104] =.16, p=.876, d=.03).
4.3 |Discussion
In Experiment 2a, we investigated whether participants could improve
learning outcomes without spending additional time studying by
watching lecture videos twice (in immediate succession) but at a faster
speed. Specifically, we hypothesized that with the time saved as a result
of watching at 2x speed, students could take advantage of the repetition
effect (improved memory performance due to repeated studying com-
pared with studying a single time, see Greene, 1989; Hintzman, 1976;
Hintzman & Block, 1971; Raajimakers, 2003) to enhance learning by
watching the videos a second time (still at 2x speed). However, although
results revealed that participants expected better learning outcomes
after only watching the videos a single time at normal speed, there were
no group differences in comprehension test performance.
The lack of comprehension benefits as a consequence of
watching the videos twice (compared with a single time) may be the
result of participants having watched the videos in immediate succes-
sion. Rather than watching the videos back-to-back, spacing learners'
multiple viewings of the videos may be a more effective study strat-
egy. Specifically, participants may be able to harness both the benefits
of repetition and the spacing effect (improved memory when studying
is spaced in time rather than in immediate succession; Bjork &
Allen, 1970; Cepeda et al., 2006; Greene, 2008; Karpicke &
Bauernschmidt, 2011) by waiting until after a delay to watch the
videos a second time. Additionally, learners could strategically distrib-
ute their study time (but keep study time constant) by rewatching lec-
tures shortly before an exam to benefit from recency effects (see
Murdock Jr., 1962). For example, if a learner were to initially watch
lectures at 2x speed and then rewatch the videos again at 2x speed
FIGURE 4 Predictions of
performance immediately before the
comprehension test and test performance
as a function of viewing schedule in
Experiment 2a. The dashed line
represents the mean performance of
participants who did not watch the
videos. Error bars reflect the SEM
6MURPHY ET AL.
immediately before an exam, this may result in better test perfor-
mance than only watching the video once initially at 1x speed and
having a longer delay before the exam.
5|EXPERIMENT 2B
To further investigate whether learners can enhance comprehension
without increasing study time, in Experiment 2b, participants again
watched lecture videos either once at normal speed or twice at 2x speed,
similar to Experiment 2a. However, participants watching the videos
twice at 2x speed spaced their viewing of the lectures and their second
viewing occurred immediately before the exam. Specifically, participants
initially either watched the videos at normal or 2x speed and after a
1-week delay, participants in the 2x speed group watched the videos a
second time immediately before taking the comprehension tests. In con-
trast, after watching the videos, the 1x speed group had a 1-week delay
before comprehension tests without any additional viewing of the lec-
tures. We expected that spacing the second viewing of the videos and
watching them immediately before the comprehension test (even at 2x
speed) would result in better learning outcomes than watching the
videos once at normal speed with a long delay before the exam, and for
participants' judgments to map on to this pattern.
5.1 |Method
5.1.1 | Participants
After exclusions, participants were 110 undergraduate students (aged 18–
35: M=21.08, SD =2.88) recruited from the UCLA Human Subjects Pool.
All participants were fluent in English and 69% were native English
speakers. Participants were tested online and received course credit for
their participation. Participants were excluded from analysis if they
admitted to cheating (e.g., looking up answers) in a post-task questionnaire
(participants were told they would still receive credit if they cheated). This
exclusion process resulted in zero exclusions. Additionally, at the end of the
study, if participants self-reported having prior expertise on either video
topic (two participants reported having expertise on appraisals and
15 reported having expertise on the Roman Empire), participants' scores on
that topic were excluded from analysis. An a priori power analysis indicated
that for a two-group test of independent means, assuming alpha =.05,
power =.80, for a two-tailed test, 90 participants would be needed to reli-
ably detect a medium effect size (d=.60).
5.1.2 | Materials and procedure
The materials were similar to Experiment 2a. Participants were ran-
domly assigned to either watch both videos once at 1x speed (n=58)
or twice at 2x speed (n=52). However, rather than taking the com-
prehension tests immediately after watching the videos, we added a
1-week delay (M=8.59 days, SD =1.29). Additionally, for partici-
pants watching the videos twice at 2x speed, rather than rewatching
the videos in immediate succession, participants watched each video
once then a second time after the 1-week delay (immediately preced-
ing the delayed comprehension test; participants were not aware that
they would watch the videos a second time). Participants also made
predictions of how they would perform after a 1-week delay after
watching the video the first time and again predicted their perfor-
mance immediately before taking the comprehension test (but after
watching the video a second time for the 2x speed group).
5.2 |Results
Predictions of performance following a delay, predictions of perfor-
mance immediately before the test, and performance as a function of
FIGURE 5 Predictions of
performance following a delay,
predictions of performance immediately
before the comprehension test, and test
performance as a function of viewing
schedule in Experiment 2b. The dashed
line represents the mean performance of
participants who did not watch the
videos. Error bars reflect the SEM
MURPHY ET AL.7
viewing schedule are shown in Figure 5. To investigate possible dif-
ferences in participants' predictions of performance following a
delay, we computed an independent samples ttest. Results rev-
ealed that after initially watching the videos at 1x speed (M=.46,
SD =.21) or 2x speed (M=.46, SD =.16), participants expected to
perform similarly (delayed predictions) on the comprehension test
regardless of viewing speed (t[108] =.01, p=.991, d< .01). How-
ever, following the 1-week delay, participants' predictions of imme-
diate performance (immediate predictions) revealed that the once
at 1x speed group (M=.36, SD =.21) expected to do worse on the
tests than participants who just rewatched the videos at 2x speed
(M=.43, SD =.19), (t[108] =2.08, p=.040, d=.40). Finally, par-
ticipants watching the videos twice at 2x speed (M=.63, SD =.14)
performed better on the comprehension tests than participants
watching the videos a single time at 1x speed (M=.55, SD =.15),
(t[108] =2.85, p=.005, d=.54).
5.3 |Discussion
In Experiment 2b, we examined whether watching lecture videos
twice at 2x speed (but spaced in time with the second viewing
occurring immediately before the exam) would result in better
learning outcomes than watching lecture videos a single time at
normal speed with a delay before the exam. In contrast to Experi-
ment 2a when videos were watched twice in immediate succession,
results revealed that watching the videos initially at 2x speed and
then again at 2x speed after a 1-week delay but immediately before
the exam resulted in better test performance than watching a single
time 1 week before the exam at normal speed. Thus, learners can
improve learning outcomes while keeping study time constant by
watching videos twice at 2x speed and strategically distributing
their viewing of lecture videos.
Again, Experiment 1 indicates that watching the videos at 2x
speed does not impair comprehension and if learners watch the
videos a second time at 2x speed immediately before an exam, they
maybeabletoharnessthebenefitsofthespacingeffect(Bjork&
Allen, 1970; Carpenter et al., 2012; Cepeda et al., 2006;
Greene, 2008; Karpicke & Bauernschmidt, 2011) and the recency
effect (Murdock Jr., 1962; see also Eitel & Scheiter, 2015). How-
ever, there is a potential confound of test delay and video speed.
Specifically, it is possible that watching a video in normal speed
once immediately before an exam could have a similar benefit as
watching at double speed immediately before an exam. Still, in
Experiment 1, there were no significant differences in performance
on the immediate tests between participants watching at normal
and double speed. Thus, watching videos twice at double speed
(withviewingsspacedintimeandthesecondviewingoccurring
immediately before an exam) resulted in enhanced performance but
a condition whereby participants watch the video once at normal
speed immediately before the test may serve as a better compari-
son group and may better represent how students actually prepare
for exams.
6|EXPERIMENT 3A
In Experiment 3a, we investigated whether watching lecture videos
multiple times but at various speeds could improve learning outcomes.
Specifically, participants either watched lecture videos initially at nor-
mal speed before watching again but at a faster speed (2x) or watched
initially at a faster speed (2x) before watching a second time at normal
speed. We were unsure as to which study schedule would lead to bet-
ter performance as there are potential benefits to each study sched-
ule. For example, it is possible that watching initially at a faster speed
may prime memory for the information (see Baddeley & Hitch, 1993),
making it easier to encode during the second viewing of the video.
Specifically, having already been exposed to the information, learners
may be more prepared to engage in elaborative encoding strategies to
better remember the information on their second viewing. Alterna-
tively, watching at normal speed initially and then again at a faster
speed may result in retrieval practice benefits (see Bjork, 1988;
Roediger III & Butler, 2011) whereby after having already encoded
the information, a second exposure to the content serves as an oppor-
tunity for learners to practice recalling the information, leading to
better retention and test performance. However, both study sched-
ules could potentially benefit from priming and/or retrieval practice
and it remains unclear how different viewing schedules impact
comprehension.
6.1 |Method
6.1.1 | Participants
After exclusions, participants were 108 undergraduate students
(aged 18–27: M=20.42, SD =1.62) recruited from the UCLA
Human Subjects Pool. All participants were fluent in English and
61% were native English speakers. Participants were tested online
and received course credit for their participation. Participants were
excluded from analysis if they admitted to cheating (e.g., looking up
answers) in a post-task questionnaire (participants were told they
would still receive credit if they cheated). This exclusion process
resulted in two exclusions. Additionally, at the end of the study, if
participants self-reported having prior expertise on either video
topic (two participants reported having expertise on appraisals and
23 reported having expertise on the Roman Empire), participants'
scores on that topic were excluded from analysis. An a priori power
analysis indicated that for a two-group test of independent means,
assuming alpha =.05, power =.80, for a two-tailed test, 90 partici-
pants would be needed to reliably detect a medium effect
size (d=.60).
6.1.2 | Materials and procedure
The materials were similar to Experiment 1. Participants watched each
video twice but were randomly assigned to either watch the videos
8MURPHY ET AL.
first at 1x speed then again at 2x speed (n=57) or first at 2x speed
then again at 1x speed (n=51). After watching the video on real
estate appraisals twice in immediate succession (participants were not
aware that they would watch the video a second time), participants
predicted their performance on the comprehension test. Participants
then completed the comprehension test and repeated this procedure
for the Roman Empire video. At the conclusion of the task, we asked
participants which study schedule they think is better for learning.
6.2 |Results
Predictions and performance as a function of viewing schedule are
shown in Figure 6. To investigate potential differences in participants'
predictions of performance, we computed an independent samples
ttest. Results revealed that participants watching the videos at 1x
speed then again at 2x speed expected to perform similarly on the
comprehension test (M=.58, SD =.18) as participants watching at 2x
speed then again at 1x speed (M=.55, SD =.19), (t[106] =.86,
p=.393, d=.17). Additionally, an independent samples ttest on test
performance did not reveal differences in comprehension between
participants watching the videos at 1x speed then again at 2x speed
(M=.65, SD =.19) and participants watching the videos at 2x speed
then again at 1x speed (M=.60, SD =.18), (t[106] =1.31,
p=.195, d=.25).
Although the two study schedules did not result in different
predictions of performance or comprehension, on the post-task
questionnaire, most participants (76%) indicated that they believed
that watching first at normal speed then rewatching at 2x speed is
better for learning than watching first at 2x speed before
rewatching at normal speed (24%); a Chi-square goodness of fit
test indicated that the frequency of these answer choices differed
(χ
2
[1] =29.04, p< .001).
6.3 |Discussion
In Experiment 3a, participants watched each video twice: once at nor-
mal speed and once at 2x speed. However, results revealed that the
order of watching the videos (i.e., fast then slow or slow then fast) did
not affect comprehension, indicating that either (1) the benefits of
priming and retrieval practice were similar or (2) learners may have
not benefitted from priming when watching first at 2x speed then at
1x speed and not benefitted from retrieval practice when watching
first at 1x speed then at 2x speed. Considering this second explana-
tion, although a second viewing may serve as an informal test of the
information encoded from the initial viewing at normal speed, simply
watching lecture videos may be too passive (i.e., participants are not
explicitly instructed to retrieve any information during the second
viewing) to result in performance benefits. Rather, more active forms
of retrieval practice such as comprehension questions embedded
within the video may be necessary to harness any benefits of retrieval
practice. Thus, although students may prefer certain study schedules
or techniques, there are instances where their beliefs about self-
regulated learning do not enhance learning outcomes (see
Azevedo, 2005).
7|EXPERIMENT 3B
In Experiment 3a, watching the lectures first at 1x speed then again at
2x speed or vice versa did not impact predictions of performance or
comprehension. Despite not observing any group differences on an
immediate comprehension test, spacing learners' viewing of the
videos may result in better performance. In Experiment 3b, we investi-
gated how delayed repetitions of varying playback speeds impacts
comprehension. Specifically, rather than watching the videos in imme-
diate succession, the second viewing of the videos occurred after a
FIGURE 6 Predictions of
performance immediately before the
comprehension test and test performance
as a function of viewing schedule in
Experiment 3a. The dashed line
represents the mean performance of
participants who did not watch the
videos. Error bars reflect the SEM
MURPHY ET AL.9
1-week delay. Similar to Experiment 3a, we were unsure as to how
the different study schedules would impact learners' predictions and
performance.
7.1 |Method
7.1.1 | Participants
After exclusions, participants were 113 undergraduate students (aged 18–
35: M=20.87, SD =2.92) recruited from theUCLAHumanSubjects
Pool. All participants were fluent in English and 66% were native English
speakers. Participants were tested online and received course credit for
their participation. Participants were excluded from analysis if they admit-
ted to cheating (e.g., looking up answers) in a post-task questionnaire (par-
ticipants were told they would still receive credit if they cheated). This
exclusion process resulted in one exclusion. Additionally, at the end of the
study, if participants self-reported having prior expertise on either video
topic (five participants reported having expertise on appraisals and
15 reported having expertise on the Roman Empire), participants' scores
on that topic were excluded from analysis. An a priori power analysis indi-
cated that for a two-group test of independent means, assuming
alpha =.05, power =.80, for a two-tailed test, 90 participants would be
needed to reliably detect a medium effect size (d=.60).
7.1.2 | Materials and procedure
The materials were similar to Experiment 3a. Participants watched
each video twice but were randomly assigned to either watch the
videos first at 1x speed then again at 2x speed (n=65) or first at 2x
speed then again at 1x speed (n=48). However, participants' second
viewing of the videos occurred after a 1-week delay (M=9.44 days,
SD =2.35). After initially watching each video (at either 1x or 2x
speed), participants predicted their performance following a 1-week
delay (participants were not aware that they would watch the video a
second time). After the 1-week delay, participants watched each video
again (at either 1x or 2x speed) and predicted their performance
before taking the comprehension test. At the conclusion of the task,
we asked participants which study schedule they think is better for
learning.
7.2 |Results
Predictions and performance as a function of viewing schedule are
shown in Figure 7. To investigate potential differences in participants'
predictions of delayed performance, we computed an independent
samples ttest. Results revealed that participants watching the videos
at 1x speed then again at 2x speed expected to perform similarly on
the delayed comprehension test (M=.48, SD =.21) as participants
watching at 2x speed then again at 1x speed (M=.48, SD =.20), (t
[111] =.09, p=.928, d=.02). Similarly, an independent samples
ttest on participants' predictions after rewatching the videos a week
later revealed that participants watching the videos at 1x speed then
again at 2x speed expected to perform similarly on the delayed com-
prehension test (M=.46, SD =.23) as participants watching at 2x
speed then again at 1x speed (M=.48, SD =.18), (t[110.7] =.52,
p=.602, d=.10). Finally, an independent samples ttest on test per-
formance did not reveal differences on the delayed comprehension
test between participants watching the videos at 1x speed then again
at 2x speed (M=.64, SD =.18) and participants watching the videos
at 2x speed then again at 1x speed (M=.62, SD =.17), (t[111] =.45,
p=.653, d=.09).
Additionally, consistent with Experiment 3a, on the post-task ques-
tionnaire, most participants (76%) indicated that they believed that
watching first at normal speed then rewatching at 2x speed is better for
learning than watching first at 2x speed before rewatching at normal
FIGURE 7 Predictions of
performance following a delay,
predictions of performance immediately
before the comprehension test, and test
performance as a function of viewing
schedule in Experiment 3b. The dashed
line represents the mean performance of
participants who did not watch the
videos. Error bars reflect the SEM
10 MURPHY ET AL.
speed (24%); a Chi-square goodness of fit test indicated that the fre-
quency of these answer choices differed (χ
2
[1] =30.81, p< .001).
7.3 |Discussion
In Experiment 3b, participants again watched each video twice:
once at normal speed and once at 2x speed. However, participants'
second viewing session occurred after a 1-week delay. Results rev-
ealed that the order of watching the videos (i.e., fast then slow or
slow then fast) did not affect performance, similar to Experiment
3a. Together, Experiment 3 suggests that if learners watch lecture
videos twice at different speeds, the order of viewing (e.g., 1x
speed before 2x speed or 2x speed then 1x speed) does not seem
to impact comprehension.
8|GENERAL DISCUSSION
Video streaming platforms (e.g., YouTube) often allow users to manip-
ulate the playback speed of videos, allowing up to 2x faster consump-
tion of a video. In addition to streaming media content, students often
manipulate the speed of asynchronous lecture videos. For example,
we surveyed 123 undergraduate students and 85% reported watching
lecture videos at quicker than normal speeds (see Figure 3a). How-
ever, 91% of students reported that they believed watching at normal
or slightly faster than normal (1.5x) as opposed to faster speeds (2x,
2.5x) was best for learning (see Figure 3b). Thus, understanding how
video speed affects short and long-term comprehension is essential to
ensure that people employ the most efficient and effective techniques
to successfully learn new information.
Previous work on the effect of video speed on learning has
yielded mixed results. For example, there is some evidence that
increasing the speed of videos leads to preserved or enhanced com-
prehension (Lang et al., 2020; Nagahama & Morita, 2017; Wilson
et al., 2018) while other work suggests that increased speed impairs
comprehension (Foulke & Sticht, 1969; Song et al., 2018; Vemuri
et al., 2004). To further examine the impact of lecture video speed on
comprehension, in Experiment 1, we tested learners' immediate and
delayed (1 week) comprehension after watching videos at either 1x,
1.5x, 2x, or 2.5x speed. As predicted by participants, results revealed
that video speed had little effect on both immediate and delayed com-
prehension such that participants only showed comprehension defi-
cits when watching at 2.5x speed compared with 1x speed. Such
learning impairments are consistent with the Cognitive Load Theory
(see Sweller, 1988, 1989) and the Cognitive Theory of Multimedia
Learning (Mayer, 2002) such that when the material is presented at
faster than 2x speed, the rate of presentation results in a cognitive
load that exceeds learners' limited cognitive resources. However,
because there appear to be minimal costs incurred by increasing video
playback speed up to 2x speed, it is possible that faster presentation
speeds do not overly tax working memory as long as the speeds do
not exceed 2x speed.
Since the present findings suggest that learners can watch videos
at up to 2x speed without accompanying performance deficits, stu-
dents may be able to use the saved time advantageously. Specifically,
students may be able to harness the memory benefits of repetition
(see Greene, 1989; Hintzman, 1976; Hintzman & Block, 1971;
Raajimakers, 2003) by watching videos twice at 2x speed without
spending additional time studying. In Experiment 2, we investigated
whether watching a video twice at 2x speed rather than a single time
at normal speed resulted in better learning outcomes. In Experiment
2a, despite participants watching once at 1x speed predicting better
performance than participants watching twice at 2x speed
(in immediate succession), the two groups performed similarly on the
comprehension test.
However, in Experiment 2b, participants either watched the
videos initially at 1x speed or 2x speed but following a 1-week delay,
participants who watched at 1x speed took the comprehension test
while participants who initially watched at 2x speed rewatched the
videos before taking the comprehension test. After the initial viewing
session, both groups predicted similar performance but following the
1-week delay and some participants rewatching the videos at 2x
speed, participants watching the videos twice at 2x speed expected to
perform better than participants watching once at 1x speed. Mirroring
the latter predictions, when the study sessions were spaced in time
(i.e., when there was a 1-week delay between encoding sessions) and
the second 2x viewing occurred immediately before the test, watching
the videos twice at 2x speed led to better performance. Thus, stu-
dents may be able to study more efficiently and enhance learning out-
comes by watching their asynchronous lectures initially at 2x speed
and again at 2x speed immediately before the test rather than
watching a single time at 1x speed with a long delay before the exam.
Again, superior exam performance only occurred when the twice at
2x speed group's second viewing occurred shortly before the exam.
Thus, learners may be able to strategically distribute their study time to
increase performance by watching lecture videos twice at 2x speed but
watching for the second time shortly before an exam. For example, in a
standard college course, there may be around 10 h of lecture video cov-
ered on a midterm exam. If students watch the lectures as they are
released each week, the time between viewing the lectures and being
tested on the material will be relatively long. Additionally, rewatching all
10 h of content shortly before the exam may not be feasible. However,
if students watch the lectures at 2x speed as they are released, then
spend just 5 h rewatching the videos shortly before the exam, they may
be able to benefit from this shorter retention interval without spending
extra time on the videos. Future work should examine various retention
intervals between encoding sessions and comprehension tests to better
understand this effect and delineate under what conditions watching
videos twice at 2x speed improves performance to a greater degree than
watching the videos once at 1x speed.
Lastly, to investigate whether learners can enhance the benefits
of repetition, participants in Experiment 3 watched each video twice:
either initially at normal speed and then again at 2x speed or initially
at 2x speed and then again at normal speed. However, both predic-
tions and performance did not differ as a function of study schedule
MURPHY ET AL.11
whether participants watched the videos in immediate succession or
with a 1-week delay between viewings. Thus, Experiment 3 indicates
that watching lecture videos multiple times at different speeds may
not be an effective study strategy to enhance comprehension.
Collectively, the present experiments indicate that increased
video speed (up to 2x) does not negatively impact learning outcomes
and watching at faster speeds can be a more efficient use of study
time. Thus, as long as to-be-remembered information can be effec-
tively perceived and encoded, learning outcomes may not be affected
by playback speed. However, previous work has indicated that speech
comprehension begins to decline at around 275 words per minute
(Foulke & Sticht, 1969; see also Goldhaber, 1970; Pastore &
Ritzhaupt, 2015; Vemuri et al., 2004) and the videos in the current
study exceeded this threshold when played at 2x speed. Although the
elevated speech rates at 2x speed may initially be less comprehensible
to students, researchers have been able to train participants to under-
stand speech at rates up to 475 WPM (Orr et al., 1965). Therefore,
with practice, higher rates of speech may not be completely incom-
prehensible and since 85% of students reported watching lecture
videos at quicker than normal speeds (see Figure 3a), they may be bet-
ter able to process the material as a result of experience.
Additionally, although the audible material presented by an instruc-
tor is impacted by playback speed in terms of speech rates, the visual
properties of an instructor's PowerPoint slides containing still images or
bulleted information should be resistant to video speed, allowing this
information to be effectively encoded if presented for sufficient time.
However, if information in a lecture video is presented only very briefly
or as an animation or moving graphic, increased speeds may decrease
the perceptibility of visual information (but see Fischer et al., 2008).
Despite having less time to encode accompanying visually presented
information at faster video speeds, the shorter durations may make it
easier for participants to sustain focused attention throughout the learn-
ing period (see Guo et al., 2014). Thus, even if the speech rates of a video
surpass the rate of what is typically perceptible, participants may still be
able to encode visually presented information (see Pastore &
Ritzhaupt, 2015) and benefit from a shorter duration with which they
need to allocate attentional resources.
The present findings provide preliminary evidence of how to
study more efficiently and illustrate that participants seem generally
aware of the effects of video speed on immediate and delayed com-
prehension as well as the benefits of rewatching lectures shortly
before the exam. However, participants' predictions prior to taking
any examinations seemed generally underconfident in relation to how
much they ended up remembering. Furthermore, learners may some-
times prefer study schedules that do not enhance performance and
future work should further examine how video speed influences
metacognitive monitoring of learning. Future work will also benefit
from investigating the factors that likely affect the efficacy of remote
learning beyond video speed. The ability to pause and take notes,
amount of audiovisual overlap, complexity or difficulty of content,
presence of subtitles, video length, embedded questions within a
video, and instructor fluency are just some of the factors that should
be further examined to bolster students' success with remote learning.
Additionally, there may be boundary conditions to the effects of video
speed on comprehension such as complexity or difficulty of the topic,
audiovisual overlap, and instructor fluency. Nevertheless, in the pre-
sent study, using two videos on different topics with different instruc-
tors and collapsing results across these topics may increase the
generalizability of the findings.
In sum, remote learning has become ubiquitous in recent times and
asynchronous learning formats may allow for more efficient learning.
Specifically, the current study revealed that there are minimal costs
incurred as a result of watching pre-recorded lectures at up to 2x speed
and if learners watch these asynchronous lectures multiple times (per-
haps rewatching the videos shortly before an exam), learning outcomes
can be enhanced. Thus, remote learning may offer students the opportu-
nity to learn both more effectively and more efficiently.
CONFLICT OF INTEREST
The authors certify that they have no affiliations with or involvement
in any organization or entity with any financial or non-financial inter-
est in the subject matter or materials discussed in this manuscript.
ENDNOTES
1
This sample was not preregistered.
2
A one-sample ttest revealed that participants watching at 2.5x speed
performed better on both the immediate (t[56] =7.00, p< .001, d=.93)
and delayed comprehension tests (t[48] =4.01, p< .001, d=.57) com-
pared with control participants who did not watch the videos.
DATA AVAILABILITY STATEMENT
The experiments reported in this article were formally preregistered
and the stimuli, data, and analysis code have been made available
on the Open Science Framework https://osf.io/zqg4k/?view_only=
c95e34ff910b41ebbaa5d86b7002881a.
ORCID
Dillon H. Murphy https://orcid.org/0000-0002-5604-3494
Kara M. Hoover https://orcid.org/0000-0001-8396-4734
Karina Agadzhanyan https://orcid.org/0000-0002-1938-6068
Jesse C. Kuehn https://orcid.org/0000-0002-3703-2431
Alan D. Castel https://orcid.org/0000-0003-1965-8227
REFERENCES
Azevedo, R. (2005). Using hypermedia as a metacognitive tool for
enhancing student learning? The role of self-regulated learning. Educa-
tional Psychologist,40(4), 199–209. https://doi.org/10.1207/
s15326985ep4004_2
Baddeley, A. D., & Hitch, G. (1974). Working memory. In Psychology of
learning and motivation (Vol. 8, pp. 47–89). Academic press. https://
doi.org/10.1016/S0079-7421(08)60452-1
Baddeley, A. D., & Hitch, G. (1993). The recency effect: Implicit learning
with explicit retrieval? Memory & Cognition,21(2), 146–155. https://
doi.org/10.3758/BF03202726
Barbour, M. K. (2013). The landscape of K-12 online learning: Examining
what is known. In M. G. Moore (Ed.), Handbook of distance education
(3rd ed., pp. 574–593). Routledge.
12 MURPHY ET AL.
Belt, E. S., & Lowenthal, P. R. (2021). Video use in online and blended
courses: A qualitative synthesis. Distance Education,42, 410–440.
https://doi.org/10.1080/01587919.2021.1954882
Bjork, R. A. (1988). Retrieval practice and the maintenance of knowledge.
In M. M. Gruneberg, P. E. Morris, & R. N. Sykes (Eds.), Practical aspects
of memory: Current research and issues, Vol. 1. Memory in everyday life
(p. 396–401). John Wiley & Sons.
Bjork, R. A., & Allen, T. W. (1970). The spacing effect: Consolidation or dif-
ferential encoding? Journal of Verbal Learning and Verbal Behavior,9(5),
567–572. https://doi.org/10.1016/S0022-5371(70)80103-7
Bjork, R. A., Dunlosky, J., & Kornell, N. (2013). Self-regulated learning:
Beliefs, techniques, and illusions. Annual Review of Psychology,64(1),
417–444. https://doi.org/10.1146/annurev-psych-113011-143823
Boekaerts, M. (1997). Self-regulated learning: A new concept embraced by
researchers, policy makers, educators, teachers, and students. Learning and
Instruction,7(2), 161–186. https://doi.org/10.1016/S0959-4752(96)
00015-1
Breslow,L.,Pritchard,D.E.,Deboer,J.,Stump,G.S.,Ho,A.D.,&Seaton,D.T.
(2013). Studying learning in the worldwide classroom: Research into edX's
first MOOC. Research and Practice in Assessment,8,13–25.
Cardall, S., Krupat, E., & Ulrich, M. (2008). Live lecture versus video-recorded
lecture: Are students voting with their feet? A cademic Medicine,83(12),
1174–1178. https://doi.org/10.1097/ACM.0b013e31818c6902
Carpenter, S. K., Cepeda, N. J., Rohrer, D., Kang, S. H., & Pashler, H.
(2012). Using spacing to enhance diverse forms of learning: Review of
recent research and implications for instruction. Educational Psychology
Review,24, 369–378. https://doi.org/10.1007/s10648-012-9205-z
Cepeda, N. J., Pashler, H., Vul, E., Wixted, J. T., & Rohrer, D. (2006). Dis-
tributed practice in verbal recall tasks: A review and quantitative syn-
thesis. Psychological Bulletin,132(3), 354–380. https://doi.org/10.
1037/0033-2909.132.3.354
Cermak, L. S., Verfaellie, L. S., Mather, M., & Chase, K. (1996). Effect of
spaced repetitions on amnesia patients' recall and recognition perfor-
mance. Neuropsychology,10(2), 219–227. https://doi.org/10.1037/
0894-4105.10.2.219
Cowan, N. (2010). The magical mystery four: How is working memory
capacity limited, and why? Current Directions in Psychological Science,
19(1), 51–57. https://doi.org/10.1177/0963721409359277
Dunlosky, J., Mueller, M. L., & Thiede, K. W. (2016). Methodology for
investigating human metamemory: Problems and pitfalls. In J.
Dunlosky & S. K. Tauber (Eds.), Oxford library of psychology. The Oxford
handbook of metamemory (pp. 23–37). Oxford University Press.
Eitel, A., & Scheiter, K. (2015). Picture or text first? Explaining
sequence effects when learning with pictures and text. Educational
Psychology Review,27, 153–180. https://doi.org/10.1007/s10648-
014-9264-4
Fischer, S., Lowe, R. K., & Schwan, S. (2008). Effects of presentation speed
of a dynamic visualization on the understanding of a mechanical sys-
tem. Applied Cognitive Psychology,22(8), 1126–1141. https://doi.org/
10.1002/acp.1426
Foulke, E., & Sticht, T. G. (1969). Review of research on the intelligibility
and comprehension of accelerated speech. Psychological Bulletin,72(1),
50–62. https://doi.org/10.1037/h0027575
Gilchrist, A. L., Cowan, N., & Naveh-Benjamin, M. (2008). Working mem-
ory capacity for spoken sentences decreases with adult ageing: Recall
of fewer but not smaller chunks in older adults. Memory,16(7), 773–
787. https://doi.org/10.1080/09658210802261124
Goldhaber, G. M. (1970). Listener comprehension of compressed speech
as a function of the academic grade level of the subjects. Journal of
Communication,20(2), 167–173. https://doi.org/10.1111/j.1460-
2466.1970.tb00874.x
Greene, R. L. (1989). Spacing effects in memory: Evidence for a two-process
account. Journal of Experimental Psychology: Learning, Memory, and Cogni-
tion,15(3), 371–377. https://doi.org/10.1037/0278-7393.19.1.159
Greene, R. L. (2008). Repetition and spacing effects. In H. L. Roediger (Ed.),
Learning and memory: A comprehensive reference. Cognitive psychology
of memory (Vol. 2, pp. 65–78). Elsevier.
Guo, P. J., Kim, J., & Rubin, R. (2014). How video production affects stu-
dent engagement: An empirical study of MOOC videos. In Proc. L&S
2014 (pp. 41–50). ACM Press.
Hintzman,D.L.(1976).Repetitionandmemory.Psychology of Learning and
Motivation,10,47–91. https://doi.org/10.1016/S0079-7421(08)60464-8
Hintzman, D. L., & Block, R. A. (1971). Repetition and memory: Evidence
for a multiple-trace hypothesis. Journal of Experimental Psychology,
88(3), 297–306. https://doi.org/10.1037/h0030907
Karpicke, J. D., & Bauernschmidt, A. (2011). Spaced retrieval: Absolute
spacing enhances learning regardless of relative spacing. Journal of
Experimental Psychology: Learning, Memory, and Cognition,37(5), 1250–
1257. https://doi.org/10.1037/a0023436
Koriat, A., Bjork, R. A., Sheffer, L., & Bar, S. K. (2004). Predicting one's own
forgetting: The role of experience-based and theory-based processes.
Journal of Experimental Psychology: General,133, 643–656. https://doi.
org/10.1037/0096-3445.133.4.643
Kornell, N., Rhodes, M. G., Castel, A. D., & Tauber, S. K. (2011). The ease-
of-processing heuristic and the stability bias: Dissociating memory,
memory beliefs, and memory judgments. Psychological Science,22,
787–794. https://doi.org/10.1177/0956797611407929
Lang, D., Chen, G., Mirzaei, K., & Paepcke, A. (2020). Is faster better?: A
study of video playback speed. In Proceedings of the tenth international
conference on learning analytics & knowledge (pp. 260–269). ACM.
https://doi.org/10.1145/3375462.3375466
Leahy, W., & Sweller, J. (2011). Cognitive load theory, modality of presen-
tation and the transient information effect. Applied Cognitive Psychol-
ogy,25, 943–951. https://doi.org/10.1002/acp.1787
Mayer, R. E. (2002). Multimedia learning. In Psychology of learning and
motivation (Vol. 41, pp. 85–139). Academic Press.
Miller, G. A. (1956). The magical number seven, plus or minus two: Some
limits on our capacity for processing information. Psychological Review,
63(2), 81–97. https://doi.org/10.1037/h0043158
Murdock, B. B., Jr. (1962). The serial position effect of free recall. Journal
of Experimental Psychology,64(5), 482–488. https://doi.org/10.1037/
h0045106
Nagahama, T., & Morita, Y. (2017). Effect analysis of playback speed for
lecture video including instructor images. International Journal for Edu-
cational Media and Technology,11(1), 50–58.
Nelson, T. O. (1996). Consciousness and metacognition. American Psychol-
ogist,51, 102–116.
Nelson, T. O., & Narens, L. (1990). Metamemory: A theoretical framework
and some new findings. In G. H. Bower (Ed.), The psychology of learning
and motivation (pp. 125–173). Academic Press.
Orr, D. B., Friedman, H. L., & Williams, J. C. C. (1965). Trainability of listen-
ing comprehension of speeded discourse. Journal of Educational Psy-
chology,56(3), 148–156. https://doi.org/10.1037/h0021987
Paas, F., Renkl, A., & Sweller, J. (2003). Cognitive load theory and instruc-
tional design: Recent developments. Educational Psychologist,38(1), 1–
4. https://doi.org/10.1207/S15326985EP3801_1
Paas, F., Renkl, A., & Sweller, J. (2004). Cognitive load theory: Instructional
implications of the interaction between information structures and
cognitive architecture. Instructional Science,32(1/2), 1–8.
Panadero, E. (2017). A review of self-regulated learning: Six models and
four directions for research. Frontiers in Psychology,8,1–8. https://doi.
org/10.3389/fpsyg.2017.00422
Pastore, R., & Ritzhaupt, A. D. (2015). Using time-compression to make
multimedia learning more efficient: Current research and practice.
TechTrends,59,66–74. https://doi.org/10.1007/s11528-015-0841-2
Peelle, J. E., & Davis, M. H. (2012). Neural oscillations carry speech rhythm
through to comprehension. Frontiers in Psychology,3, 320. https://doi.
org/10.3389/fpsyg.2012.00320
MURPHY ET AL.13
Pokhrel, S., & Chhetri, R. (2021). A literature review on impact of COVID-
19 pandemic on teaching and learning. Higher Education for the Future,
8, 133–141. https://doi.org/10.1177/2347631120983481
Raajimakers, J. G. W. (2003). Spacing and repetition effects in human
memory: Application of the SAM model. Cognitive Science,27(3), 431–
452. https://doi.org/10.1207/s15516709cog2703_5
Roediger, H. L., III, & Butler, A. C. (2011). The critical role of retrieval prac-
tice in long-term retention. Trends in Cognitive Sciences,15,20–27.
https://doi.org/10.1016/j.tics.2010.09.003
Sinha, T., Jermann, P., Li, N., & Dillenbourg, P. (2014). Your click decides
your fate: Inferring information processing and attrition behavior from
MOOC video clickstream interactions. In Proceedings of the 2014 con-
ference on empirical methods in natural language processing (EMNLP)
(pp. 3–14). Association for Computational Linguistics. https://doi.org/
10.3115/v1/w14-4102
Song, K., Chakraborty, A., Dawson, M., Dugan, A., Adkins, B., & Doty, C.
(2018). Does the podcast video playback speed affect comprehension
for novel curriculum delivery? A randomized trial. The Western Journal
of Emergency Medicine,19, 101–105. https://doi.org/10.5811/
westjem.2017.10.36027
Sweller, J. (1988). Cognitive load during problem solving: Effects on learn-
ing. Cognitive Science,12(2), 257–285. https://doi.org/10.1016/0364-
0213(88)90023-7
Sweller, J. (1989). Cognitive technology: Some procedures for facilitating
learning and problem solving in mathematics and science. Journal of
Educational Psychology,81(4), 457–466. https://doi.org/10.1037/
0022-0663.81.4.457
Sweller, J., Ayres, P., & Kalyuga, S. (2011). Measuring cognitive load. In
Cognitive load theory (pp. 71–85). Springer. https://doi.org/10.1007/
978-1-4419-8126-4_6
Sweller, J., van Merriënboer, J. J., & Paas, F. G. (1998). Cognitive architec-
ture and instructional design. Educational Psychology Review,10(3),
251–296. https://doi.org/10.1023/A:1022193728205
Thiede, K. W., & De Bruin, A. B. H. (2017). Self-regulated learning in read-
ing. In D. H. Schunk & J. A. Greene (Eds.), Handbook of self-regulation
of learning and performance (2nd ed., pp. 109–123). Routledge. https://
doi.org/10.4324/9781315697048-8
van Merriënboer, J. J., & Sweller, J. (2005). Cognitive load theory and com-
plex learning: Recent developments and future directions. Educational
Psychology Review,17(2), 147–177. https://doi.org/10.1007/s10648-
005-3951-0
Vemuri, S., DeCamp, P., Bender, W., & Schmandt, C. (2004).
Improving speech playback using time-compression and speech recog-
nition. In Proceedings of SIGCHI conference human factors in computing
systems (pp. 295–302). ACM. https://doi.org/10.1145/985692.
985730
Wilson, K. E., Martin, L., Smilek, D., & Risko, E. F. (2018). The benefits and
costs of speed watching video lectures. Scholarship of Teaching and
Learning in Psychology,4(4), 243–257. https://doi.org/10.1037/
stl0000127
Wong, J., Baars, M., Davis, D., Van Der Zee, T., Houben, G., & Paas, F.
(2019). Supporting self-regulated learning in online learning environ-
ments and MOOCs: A systematic review. International Journal of
Human Computer Interaction,35(4–5), 356–373. https://doi.org/10.
1080/10447318.2018.1543084
Zimmerman, B. J. (1989). A social cognitive view of self-regulated aca-
demic learning. Journal of Educational Psychology,81(3), 329–339.
https://doi.org/10.1037/0022-0663.81.3.329
Zimmerman, B. J. (1990). Self-regulating academic learning and achieve-
ment: The emergence of a social cognitive perspective. Educational
Psychology Review,2(2), 173–201. https://doi.org/10.1007/
bf01322178
How to cite this article: Murphy, D. H., Hoover, K. M.,
Agadzhanyan, K., Kuehn, J. C., & Castel, A. D. (2021). Learning
in double time: The effect of lecture video speed on
immediate and delayed comprehension. Applied Cognitive
Psychology,1–14. https://doi.org/10.1002/acp.3899
14 MURPHY ET AL.