Professional Documents
Culture Documents
An AR Learning Approch To Enhancing Reading Science
An AR Learning Approch To Enhancing Reading Science
doi:10.1111/bjet.12716
Abstract
Reading has been regarded as a medium for learning science, revealing the importance
of enhancing learners’ reading competence in science education. The critical features of
science texts are their multiple representations, such as text and visual elements, which
assist the representation of science concepts. A multimedia learning environment can
present relevant materials in various formats and help students to process the materials
in meaningful ways, for example, by integrating learning materials with relevant prior
concepts, and organizing them into a consistent and coherent cognitive structure.
However, some issues with multimedia instructional design have been proposed, such
as students’ cognitive load and learning motivations. In this study, an augmented
reality-based science learning system was developed based on the contiguity principle
of multimedia learning in order to promote students’ science learning. Moreover, an
experiment was conducted on a natural science course in an elementary school to
assess the effectiveness of the implemented system on students’ learning. The
experimental results display that the students learning with this approach found made
significant gains in their learning achievements and motivations compared to those
learning science with conventional multimedia science learning; moreover, their
perceptions of extraneous cognitive load were significantly reduced during the learning
activity.
Introduction
Several previous studies have advanced the significance of reading competence in science ed-
ucation for conceptual and theoretical comprehension (Yang, Chang, Chen, & Chen, 2016).
Researchers have asserted that readers can construct meaning by assimilating text information
with their relevant background beliefs or knowledge (Wang & Chen, 2014). Moreover, science
Practitioner Notes
What is already known about this topic
• Multimedia learning environments can offer potential affordances for facilitating
students’ concept understanding.
• The advancements in mobile technologies have enabled teachers to design rich mul-
timedia learning environments via immediately dynamic representations.
• Some challenges concerning the instructional design of multimedia have arisen,
including issues related to cognitive load.
What this paper adds
• An augmented reality-based science learning approach was developed based on the
contiguity principle of multimedia learning.
• An augmented reality-based science learning system was designed based on the im-
plemented approach.
• It was displayed that the approach improved students’ learning achievements and
motivations.
• Students’ perceptions of extraneous cognitive load were significantly reduced during
the augmented reality-based science learning activity.
Implications for practice and/or policy
• This study demonstrated the positive influence of augmented reality on a multimedia
learning situation.
• The augmented reality-based science learning approach could be a valuable refer-
ence for projecting a multimedia learning environment.
reading concerns metacognitive processes and active knowledge construction via revisiting
concepts connected to experienced authentic phenomena (Cervetti, Barber, Dorph, Pearson, &
Goldschmidt, 2012). Reading has been regarded as a medium for learning science and as being
essential to scientific works (Hung, 2014), revealing the importance of enhancing learners’ read-
ing competence in science education.
Reading competence has become increasingly crucial due to the advances in information tech-
nologies which facilitate easier access to information (Yang et al., 2016). The critical features of
science texts are the multiple representations, such as text and visual elements, which assist the
representation of science concepts (Hung, 2014). For example, Mason, Tornatora and Pluchino
(2013) asserted that visualizations play a significant role in science learning because they make
complex processes visible. Thus, it is necessary for teachers to assist students with an environ-
ment consisting of various kinds of information. Multimedia learning refers to an environment
for constructing different types of mental representations by assisting students with various kinds
of media materials, such as texts and pictures (Eitel, Scheiter, Schüler, Nyström, & Holmqvist,
2013). Moreover, the advancements in mobile technologies have enabled teachers to design
rich multimedia learning environments via immediately dynamic representations. For example,
Wang, Chen, Fang and Chou (2014) described that the increasing use of digital technologies has
supplied students with an information-driven environment to enhance their reading proficiency.
However, despite the advantages of multimedia which facilitate learning, some challenges con-
cerning the instructional design of multimedia have arisen, including issues related to cognitive
load (Knörzer, Brünken, & Park, 2016).
Literature review
Science reading could potentially effective if a student inferentially links between the relevant
knowledge and the text (Phillips & Norris, 2009). Moreover, multiple representations relate to
the reader’s prior knowledge of the science concepts to facilitate the representation and science
learning (Hung, 2014). Multimedia learning occurs when students simultaneously obtain in-
formation in all sorts of formats, such as text, pictures, animation, and video (Mayer, 1997).
Multimedia learning refers to a multistage cognitive process during which some types of cog-
nitive representation are constructed via various kinds of media materials (Eitel et al., 2013).
Thanks to the considerable advances in computer technologies, computer-based multimedia
learning environments now offer potential affordances for facilitating students’ concept under-
standing. Some previous research has explored the impacts of multimedia learning on education,
such as transfer test performance, mental effort, and higher-order thinking skills (Arslan-Ari,
2018; Chung, Cheon, & Lee, 2015). For example, She and Chen (2009) investigated the impact of
multimedia effects on science learning, and stated that the students that learned with animation
with narration outperformed those that used animation with on-screen text, in terms of their
immediate and long-term retention.
Multimedia offers a kind of effective learning environment by combining a variety of formats
such as text, pictures, animations, audio, and videos (Doolittle, Bryant, & Chittum, 2015). A mul-
timedia learning environment can present relevant materials in various formats and help stu-
dents process the materials in meaningful ways, for example, by integrating learning materials
with relevant prior concepts, and organizing them into a consistent and coherent cognitive struc-
ture. Moreover, Mayer (2003) noted that the multimedia effect produces better learning results
when it integrates pictures with words rather than just setting words in a computer-based or a
book-based environment.
Meaningful learning requires students’ engagement in considerable cognitive processing during
learning activities, but their capacity for cognitive processing is a severe restriction (Blayney et al.,
2015). This means that teachers must identify and consider the demands of cognitive load while
designing multimedia instruction. Cognitive load is referred to as a kind of multidimensional con-
struct representation of the load which is imposed on students’ cognitive systems while conduct-
ing a specific task (Paas & van Merriënboer, 1994). Cognitive load theory claims the significant
role of the interactions between a learning task and students’ limited working memory (Ayres,
2006), illustrating the assessable dimensions of mental load, mental effort, and performance.
These dimensions, respectively, correspond to three types of cognitive load, namely intrinsic,
extraneous, and germane cognitive load, which make up the total cognitive load. Intrinsic cogni-
tive load refers to an inherent component of the materials itself and to the degree of the learners’
prior knowledge or experience (Cheng, Lu, & Yang, 2015). Extraneous cognitive load is gener-
ated from that part of the instructional design which induces excess information processing and
can be modified, while germane cognitive load is considered as the amount of working memory
capacity needed to deal with intrinsic load (Leahy & Sweller, 2016).
When more working memory resources are utilized to cope with the interacting elements related
to intrinsic cognitive load, and fewer are used to deal with those regarding extraneous cognitive
load, the instruction will be more effective (Paas & Sweller, 2014). The total cognitive load can-
not exceed a student’s working memory capacity, indicating the significance of identifying the
association of total cognitive load with an instructional design. On the other hand, in Mayer’s
(2014a) study of cognitive theory of multimedia learning, he asserted that, for constructing a
coherent mental representation, humans are actively involved in cognitive processing, includ-
ing “paying attention to relevant incoming information, organizing incoming information into
a coherent cognitive structure, and integrating incoming information with other knowledge” (p.
50). Therefore, to acquire the potential positive impacts of multimedia learning, it is significant to
carefully design a learning activity that may involve students’ cognitive load.
A great deal of the previous research has illustrated various ways of properly designing multi-
media instructional activities. A study on reviewing the design of multimedia explanations con-
ducted by Mayer (2003) described the effect of multimedia instruction on learning, including
a multimedia effect, a coherence effect, a spatial contiguity effect and a personalization effect.
Among these effects, the spatial contiguity effect or principle states that simultaneously display-
ing corresponding pictures and words is better than displaying them separately in a multimedia
presentation design based on reducing extraneous cognitive load (Brunken, Plass, & Leutner,
2003). Furthermore, Türk and Erçetin (2014) designed the presentations of multimedia infor-
mation for improving students’ English vocabulary learning and reading comprehension based
on the temporal contiguity principle, finding that simultaneously displaying verbal and visual
materials can reduce students’ cognitive load and enhance their learning.
Moreover, some previous research has asserted that learners’ emotions play a crucial role in
multimedia learning (Knörzer et al., 2016). Emotional awareness when designing multimedia
learning materials can arouse students’ positive emotions which will assist the learning process
(Heidig, Müller, & Reichelt, 2015). Mayer (2014b) proposed that motivational features can help
students engage in learning processes and improve their learning if they are not overloaded with
or distracted from processing the content. As mentioned above, it is worth probing students’ moti-
vational state and their cognitive load during multimedia learning activities.
AR is referred to as a system which can integrate virtual (computer-generated) elements or
information with physical environments (Azuma et al., 2001). Azuma (1997) considered AR
as a variation of virtual environments, which consist of three characteristics: “being interac-
tive in real time,” “combining real and virtual,” and “being registered in three dimensions.”
AR can afford students with potential opportunities for linking what they observe in authentic
© 2018 British Educational Research Association
236 British Journal of Educational Technology Vol 50 No 1 2019
environments to their prior learning concepts by both visualization and interaction. For exam-
ple, Yilmaz, Kucuk, and Goktas (2017) investigated the effectiveness of AR picture books (ARPB)
on preschool children’s reading attitudes and story comprehension performance. Most of them
expressed their enjoyment while using the ARPB, and displayed high story comprehension
performance. Moreover, Yoon et al. (2017) described how AR has evidenced its advantages by
means of augmenting students’ interactions, engagement, and experiences in science learning.
Specifically, combining multiple data sources with a well-designed presentation in an AR learn-
ing environment can reduce the participants’ cognitive workload (Wei, Weng, Liu, & Wang,
2015).
AR has been widely adopted in education due to the popularization of mobile technology (Nadolny,
2017), and has demonstrated positive influences on conceptual understanding, learning experi-
ence, as well as motivation improvement (Cheng & Tsai, 2013). Furthermore, a systematic review
conducted by Ibáñez and Delgado-Kloos (2018) analyzed the 28 papers from 2010 to 2017 on
AR-based learning in science, technology, engineering, and mathematics (STEM). They asserted
that most of the studies investigated the influences of AR on students’ conceptual understand-
ing, followed by those that surveyed affect in AR-based learning. For example, Akçayır, Akçayır,
Pektaş, and Ocak (2016) probed the influence of utilizing AR technologies on students’ science
learning and found that AR can improve students’ laboratory skills as well as their positive learn-
ing attitudes.
However, few studies have probed the application of AR in education from the perspective of
multimedia learning and cognitive load. A systematic review, executed by Akçayır and Akçayır
(2017) examined the 68 research articles on AR-based educational settings in the SSCI jour-
nals, and found that promoting learning achievements was the most mentioned advantage of
AR technology. Only four of the 68 papers presented the information about cognitive load, and
contradictory results were yielded—two papers declared that it decreased cognitive load, while
the other one asserted that AR increased cognitive overload. Cheng (2017) conducted a quanti-
tative study of 153 students’ perceptions of reading an AR book, and described that the students
were aware of stronger motivation, more positive attitudes and lower cognitive load during the
AR book reading activity. He further indicated that cognitive load can predict the students’ behav-
ioral intentions when their perceptions of the factors of motivation, for example attention or
confidence, were significant. Therefore, it is crucial to probe the role of motivational factors and
cognitive load in AR-based learning environments.
Research Questions
In this study, an AR-based learning approach integrated with multimedia materials, based on
the contiguity effect of multimedia instruction, was designed for the students’ science textbook.
It is called the AR multimedia textbook (ARMTB). Moreover, an experiment was conducted
to explore the effects of the implemented approach on elementary school students’ learning
achievements and motivations. The study also probes their perceptions of cognitive load during
the learning activity. The learning effectiveness was evaluated via answering the following re-
search questions.
1. Do the students using the ARMTB approach outperform those learning science with the
conventional multimedia science learning in terms of their learning achievements?
2. Do the students using the ARMTB approach reveal higher learning motivations than those
learning science with the conventional multimedia science learning?
3. Can the ARMTB approach reduce the students’ cognitive load in comparison with the conven-
tional multimedia science learning?
© 2018 British Educational Research Association
Augmented reality-based science learning 237
Figure 3: A related picture overlaps on the textbook immediately after the scan
[Colour figure can be viewed at wileyonlinelibrary.com]
After completing the reality-based science learning, the students immediately work on the prob-
lem-based learning tasks regarding the topography being studied. During the learning tasks, indi-
vidual students will receive related prompt feedback and points from the instant response module
and the scoring module, respectively.
Figure 4: A related picture overlaps on the textbook immediately after the scan
[Colour figure can be viewed at wileyonlinelibrary.com]
Method
Participants
From two classes of fifth graders (10- or 11-year-olds) in an elementary school in Taiwan, 46 stu-
dents participated in the learning activities in this study. The two classes were allocated to be the
experimental group (n = 23) and the control group (n = 23). The experimental group conducted
the learning activity with AR-based science learning; on the other hand, the control group stud-
ied the course using conventional multimedia for science learning. All participants took natural
science courses for four periods a week and were instructed by the same teacher who has rich
experience in science teaching. Furthermore, the students were already familiar with mobile
technology, multimedia learning, and AR-based learning.
Experimental procedure
In this study, the experiment was conducted on the natural science course in an elementary
school, which aimed to enhance the students’ science reading effectiveness. Figure 5 depicts the
experimental procedure. Firstly, all participants took the pretest to evaluate their basic knowl-
edge of geography before the learning activity.
Following that, all participants implemented the learning activities for 80 minutes. During the
learning activities, the experimental group adopted the AR-based science learning (as shown in
Figure 6), while the control group utilized the conventional multimedia science learning. The
printed textbook, multimedia materials, and tasks for both groups were identical except for the
3D objects.
After completing the learning activity, all participants conducted the posttest and filled out the
questionnaires related to motivation and cognitive load. Afterward, the experiment procedure
ended in a one-on-one interview, which probed the views of six students randomly recruited from
each of the learning approach groups.
Figure 5: The procedure of the experiment for the learning activity [Colour figure can be viewed at
wileyonlinelibrary.com]
Figure 6: The scenarios of the AR-based science learning activity [Colour figure can be viewed at
wileyonlinelibrary.com]
Measuring tools
The measuring tools in this study were composed of the posttest and the surveys of learning
motivations and cognitive load.
The pretest aimed to assess the students’ fundamental knowledge of the geography before par-
ticipating in the learning activity. It comprised 20 multiple-choice items. Moreover, the posttest,
which consisted of 20 multiple-choice items, was used to evaluate what the participants had
learned during the learning activity. Both tests were scored on a scale of 0–100. To confirm the
validity and reliability of all of the items, two expert teachers with more than 10 years’ experi-
ence in teaching elementary schools’ natural science courses evaluated all items’ content valid-
ity; furthermore, via the KR20 analysis, the reliability coefficient of the pretest and the post-test,
respectively, were 0.83 and 0.79, indicating the sufficient reliability of the tests.
Keller’s ARCS Model (1987), which aims at creating and maintaining motivational environments
as well as the meaningfulness of instruction (Shellnut, Knowltion, & Savage, 1999), was uti-
lized to evaluate the students’ learning motivations in this study. The survey of learning moti-
vations was adopted from the “Instructional Materials Motivation Survey, IMMS” developed by
Keller (2010). The original IMMS comprised four dimensions, namely “Attention,” “Relevance,”
“Confidence,” and “Satisfaction.” The surveys of learning motivations consisted of 36 items
with a 5-point rating scale. Moreover, the attention dimension included 12 items, such as “These
multimedia materials caught my eye,” and “The way the learning materials are presented on
the textbook helped keep my attention.” The relevance dimension consisted of nine items. Two
example items are “There were examples, or pictures that can display for me how these learning
materials could be significant to some people” and “The content of these learning materials is
related to my interests.” The confidence dimension comprised nine items, such as “After study-
ing for this course, I am confident that I could pass an examination on the related content” and
“The good organization of the content helped me feel confident that I would learn this material.”
Furthermore, a total of six items (eg, “I enjoyed this course so much that I would like to learn
more about this topic” and “I would propose this learning approach to others”) made up the
satisfaction dimension. The Cronbach’s alpha coefficients provided by the original measurement
for the dimensions were .89, .81, .90, and .92, respectively, while the reliability coefficient of the
total scale was .96, indicating highly acceptable reliability in internal consistency of the learning
motivations survey.
The survey of cognitive load was modified from the study of Hwang, Yang, and Wang (2013),
using a 5-point Likert rating scheme. It comprised two dimensions, “mental effort” and “men-
tal load” which were, respectively, applied for displaying students’ extraneous cognitive load and
intrinsic cognitive load. Mental effort can be measured when students are executing a task, while
mental load originates from the interaction between the task and the characteristics of the learn-
ing materials. In this study, the mental effort dimension was made up of three items, such as “I
need to make a special effort to complete the learning tasks or obtain the learning achievement in
this learning activity”; on the other hand, the mental load dimension consisted of five items (e.g.,
“I felt frustrated when answering the questions in this learning activity”). The Cronbach’s alpha
coefficient values described by the original survey were .86 and .85 for the mental load dimension
and the mental effort dimension, respectively, indicating high internal consistency reliability for
assessing students’ cognitive load.
Experimental results
Learning achievement
To explore the influence of the AR-based science learning approach on students’ learning
achievements, a one-way ANCOVA was conducted. The pretest score was utilized as the covari-
ate for excluding any interference from their prior knowledge; moreover, the learning approach
and the posttest score were an independent variable and a dependent variable, respectively.
In order to confirm the reasonableness of the adoption of ANCOVA, the homogeneity test was
examined first. It was found that the assumption of homogeneity of regression was not violated
(F = 1.55, p > .05). Following that, the ANCOVA was conducted and the effect is presented
in Table 1. A significant effect of the learning approach was verified with F = 5.03 (p < .05,
η2 = 0.105), implying that significantly different posttest scores existed between the two groups
due to the different learning approaches. Moreover, the adjusted mean values and standard devi-
ation errors of the students’ posttest scores were 87.01 and 2.56 for the experimental group and
78.86 and 2.56 for the control group. This result indicates that the AR-based learning approach
can improve students’ science learning achievements. Moreover, based on the categories stated
by Cohen (1988), the effect size (η2) for the ANCOVA results of the learning approach showed a
moderate effect size with η2 = 0.105 (η2 > 0.059).
Learning motivation
As regards the students’ learning motivations, a t test was computed to investigate the influence
that the AR-based science learning approach had on the participants. As shown in Table 2, a
significant difference was confirmed between the two groups (t = 3.60; p < .001). The means
of the students’ learning motivation ratings for the experimental and the control groups were,
respectively, 4.15 (SD = 0.48) and 3.59 (SD = 0.57). It was found that the learning motivations of
the students using the AR-based science learning approach were significantly higher than those
of the students utilizing the conventional multimedia science learning. Moreover, based on the
statements described by Cohen (1988), the learning approach displayed a large effect size with
d = 1.06 (d > 0.80).
In this study, the learning motivation questionnaire comprised four dimensions, that is, atten-
tion, relevance, confidence, and satisfaction. Independent t tests were computed for evaluat-
ing the impacts of the AR-based science learning approach on the questionnaire ratings of the
four dimensions. As described in Table 3, significant differences were confirmed with t = 4.37
(p < .001) for the attention dimension, t = 2.92 (p < .01) for the relevance dimension, t = 2.80
(p < .01) for the confidence dimension, and t = 2.09 (p < .05) for the satisfaction dimension.
For the attention dimension, the means and standard deviations were 4.16 and 0.52 for the
experimental group, and 3.43 and 0.62 for the control group, indicating that the experimen-
tal group paid more attention to the learning activity than the control group did. Moreover, the
AR-based science learning approach presented a large effect size in the attention dimension with
d = 1.28 (d > 0.80). For the relevance dimension, the means were, respectively, 4.14 (SD = 0.53)
Variable and
source Group n Mean SD t d
Table 3: The t-test results of the four subscales of learning motivation for the two groups’ motivations
Variable and
source Group n Mean SD t d
and 3.67 (SD = 0.54) for the experimental group and the control group, implying that the exper-
imental group rated their perceptions of relevance to the learning activities higher than the con-
trol group did. The learning approach showed a large effect size in the relevance dimension with
d = 0.88 (d > 0.80). With regard to the confidence dimension, the means and standard deviations
were 4.10 and 0.46 for the experimental group, and 3.60 and 0.73 for the control group, indi-
cating that the experimental group was more confident after the course than the control group,
and the AR-based science learning approach had a large effect size in the confidence dimension
(d = 1.28). As regards the satisfaction dimension, the means were 4.23 (SD = 0.61) and 3.80
(SD = 0.79) for the experimental group and the control group, respectively. It was found that the
perceptions of satisfaction in the learning activities of the experimental group were higher than
those perceptions of satisfaction within the control group. Moreover, the learning approach had
a moderate effect size in the confidence dimension with d = 0.61(d > 0.5).
To sum up the effects of the learning approach on the students’ learning motivations, the
AR-based science learning approach had significant positive influences on the students’ percep-
tions of their learning motivations in four dimensions, namely attention, relevance, confidence,
and satisfaction.
Cognitive load
In this study, the survey of cognitive load was divided into two dimensions: “mental load” and
“mental effort.” The t tests were executed to probe the influences of the AR-based science learn-
ing approach on students’ intrinsic and extraneous cognitive load.
In terms of the mental load dimension (as shown in Table 4), no significant difference was found
in the two groups, with t = −1.55 (p > .05), indicating that there was no significant influence of
the AR-based science learning approach on the students’ intrinsic cognitive load. On the other
hand, a significant difference was confirmed between the two groups’ perceptions of mental effort,
with t = −2.07 (p < .05). Furthermore, the means were, respectively, 1.98 (SD = 0.71) and 2.54
(SD = 1.06) for the experimental group and for the control group, illustrating that the students who
learned with the AR-based science learning approach perceived lower extraneous cognitive load
than those who learned with conventional multimedia science learning during the learning activity.
Interviews
To acquire more detailed understanding of the participants, six students in the experimental
group (two each of the low-, medium-, and high-achieving students) participated in indepen-
dent, semi-structured interviews conducted by one researcher of this study. From the analytical
Table 4: The t test result of the two groups’ cognitive load levels
Variable and
source Group n Mean SD t d
results of the interview, it was found that the participants expressed some positions on the
ARMTB approach, including “promoting learning interest,” “presenting understandable mate-
rials,” and “inspiring learning confidence.”
From the position of “promoting learning interest,” all of the students indicated that they were
interested in the ARMTB learning activities. For example, one student revealed that, “Using AR
technology makes learning more interesting.” Another further mentioned that, “AR technology
can be utilized in diverse subjects to enhance students’ learning interests.” From the “presenting
understandable materials” perspective, five of the six participants indicated that students can
easily comprehend the abstract concepts of the science books and reduce the barriers of the read-
ing, via combining the AR technology and the textbook, and the interactions with the multime-
dia materials. For example, one student stated that, “We can understand more about the process
of Terrain Formation.” In terms of “inspiring learning confidence,” four of the six students indi-
cated that the ARMTB approach can improve their learning confidence and triggers their active
learning. For instance, one student expressed that, “The effect of the AR is excellent. I think that
my learning achievement will be improved by the AR-based learning.”
representation of their own experiences. As mentioned above, this can explain why the ARMTB
approach can effectively reduce students’ extraneous cognitive load in multimedia science learn-
ing activities, illustrating that adequately designing AR-based learning activities according to the
contiguity principle of multimedia learning has the benefit of decreasing students’ cognitive load.
As emotion plays a crucial role in multimedia learning (Knörzer et al., 2016), it is crucial to
carefully implement the motivational design for stimulating and maintaining a more motivat-
ing learning environment. The ARMTB approach can effectively promote the students’ learn-
ing motivations in terms of all four dimensions of “Attention,” “Relevance,” “Confidence,” and
“Satisfaction.” Such a positive result may originate from the characteristics of AR, for example,
augmenting students’ interactions, engagement, and experiences (Yoon et al. 2017). This com-
plies Cheng and Tsai’s (2013) assertion, illustrating that AR has a positive influence on improving
students’ learning motivations. This result also echoes those of Wei et al. (2015), who integrated
AR into the design of their study to promote students’ learning motivation to work on a general
technical creative design. Moreover, in Cheng’s (2017) quantitative survey regarding AR book
reading, the motivational factors of “Attention” and “Confidence” were deemed the crucial roles
in students’ perceived cognitive load, so as to predict their behavioral intentions. This study fur-
ther proved the significant effect of the factor of “Relevance” on decreasing students’ extrane-
ous cognitive load in the AR-based learning environments. This corresponds to the view stated
by the interviewed students, who considered the ARMTB approach as “presenting understand-
able materials” which can reduce the dilemma of reading. This could also be the reason why the
ARMTB approach is of great benefit in terms of decreasing students’ extraneous cognitive load.
The appropriate design of multimedia learning materials can enhance students’ positive emo-
tions and motivation and hence support the learning process (Heidig et al., 2015). This is one fac-
tor as to why the ARMTB approach may significantly improve students’ learning achievements
on science learning. This result is consistent with the findings described by Yoon et al. (2017),
which stated that students’ concept learning was significantly improved via interacting with the
AR technology. Furthermore, Mayer (2014b) illustrated that motivational features can stimulate
students to engage in learning processes so as to improve their learning achievement when they
are not overloaded with processing. In this study, the ARMTB approach not only promoted stu-
dents’ learning motivations but also reduced their extraneous cognitive load, indicating a valu-
able reference for projecting a multimedia learning environment.
This study illustrated the potential advantages of utilizing AR technology in science reading.
However, it should be noted that only a small-scale ARMTB learning activity was employed. In
future research, it would be worth conducting a large-scale experiment or a quantitative survey
for probing the influences of AR on science reading. Furthermore, with the advancements in
media technologies, the ways to design a rich multimedia environment are becoming ever more
diverse. The effects of multimedia learning could be evaluated by means of various advanced
techniques in further research. Moreover, this study demonstrated the positive influence of AR on
a multimedia learning situation based on the contiguity principle. More studies about the impact
of different principles of multimedia learning on students’ learning effectiveness could be investi-
gated via multiplex learning approaches in the future.
Acknowledgements
The participants of this study were protected using a serial number to replace their personal in-
formation. They were informed that participation was voluntary and would not affect the grade
of the course; moreover, they could withdraw from the study at any time.
References
Akçayır, M., & Akçayır, G. (2017). Advantages and challenges associated with augmented reality for edu-
cation: A systematic review of the literature. Educational Research Review, 20, 1–11.
Akçayır, M., Akçayır, G., Pektaş, H. M., & Ocak, M. A. (2016). Augmented reality in science laboratories:
The effects of augmented reality on university students’ laboratory skills and attitudes toward science
laboratories. Computers in Human Behavior, 57, 334–342.
Arslan-Ari, I. (2018). Learning from instructional animations: How does prior knowledge mediate the
effect of visual cues? Journal of Computer Assisted Learning, 34(2), 140–149.
Ayres, P. (2006). Using subjective measures to detect variations of intrinsic cognitive load within prob-
lems. Learning and Instruction, 16(5), 389–400.
Azuma, R. T. (1997). A survey of augmented reality. Presence: Teleoperators and Virtual Environments, 6(4),
355–385.
Azuma, R., Baillot, Y., Behringer, R., Feiner, S., Julier, S., & MacIntyre, B. (2001). Recent advances in aug-
mented reality. IEEE Computer Graphics and Applications, 21(6), 34–47.
Blayney, P. J., Kalyuga, S., & Sweller, J. (2015). Using cognitive load theory to tailor instruction to levels of
accounting students’ expertise. Educational Technology & Society, 18(4), 199–210.
Bower, M., Lee, M. J., & Dalgarno, B. (2017). Collaborative learning across physical and virtual worlds:
Factors supporting and constraining learners in a blended reality environment. British Journal of
Educational Technology, 48(2), 407–430.
Brunken, R., Plass, J. L., & Leutner, D. (2003). Direct measurement of cognitive load in multimedia learn-
ing. Educational Psychologist, 38(1), 53–61.
Cervetti, G. N., Barber, J., Dorph, R., Pearson, P. D., & Goldschmidt, P. G. (2012). The impact of an integrated
approach to science and literacy in elementary school classrooms. Journal of Research in Science Teaching,
49(5), 631–658.
Cheng, K. H. (2017). Reading an augmented reality book: An exploration of learners’ cognitive load, moti-
vation, and attitudes. Australasian Journal of Educational Technology, 33(4), 53–69.
Cheng, T. S., Lu, Y. C., & Yang, C. S. (2015). Using the multi-display teaching system to lower cognitive load.
Educational Technology & Society, 18(4), 128–140.
Cheng, K. H., & Tsai, C. C. (2013). Affordances of augmented reality in science learning: Suggestions for
future research. Journal of Science Education and Technology, 22(4), 449–462.
Chung, S., Cheon, J., & Lee, K. W. (2015). Emotion and multimedia learning: An investigation of the effects
of valence and arousal on different modalities in an instructional animation. Instructional Science, 43(5),
545–559.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence
Earlbaum Associates.
Doolittle, P. E., Bryant, L. H., & Chittum, J. R. (2015). Effects of degree of segmentation and learner disposi-
tion on multimedia learning. British Journal of Educational Technology, 46(6), 1333–1343.
Eitel, A., Scheiter, K., Schüler, A., Nyström, M., & Holmqvist, K. (2013). How a picture facilitates the process
of learning from text: Evidence for scaffolding. Learning and Instruction, 28, 48–63.
Heidig, S., Müller, J., & Reichelt, M. (2015). Emotional design in multimedia learning: Differentiation on
relevant design features and their effects on emotions and learning. Computers in Human Behavior, 44,
81–95.
Hung, Y. N. (2014). “What are you looking at?” an eye movement exploration in science text reading.
International Journal of Science and Mathematics Education, 12(2), 241–260.
Hwang, G. J., Yang, L. H., & Wang, S. Y. (2013). A concept map-embedded educational computer game
for improving students’ learning performance in natural science courses. Computers & Education, 69,
121–130.
Ibáñez, M. B., & Delgado-Kloos, C. (2018). Augmented reality for STEM learning: A systematic review.
Computers & Education, 123, 109–123.
Keller, J. M. (1987). Development and use of the ARCS model of instructional design. Journal of Instructional
Development, 10(3), 2–10.
Keller, J. M. (2010). Motivational design for learning and performance. New York, NY: Springer.
Knörzer, L., Brünken, R., & Park, B. (2016). Emotions and multimedia learning: The moderating role of
learner characteristics. Journal of Computer Assisted Learning, 32(6), 618–631.
Leahy, W., & Sweller, J. (2016). Cognitive load theory and the effects of transient information on the modal-
ity effect. Instructional Science, 44(1), 107–123.
Mason, L., Tornatora, M. C., & Pluchino, P. (2013). Do fourth graders integrate text and picture in process-
ing and learning from an illustrated science text? Evidence from eye-movement patterns. Computers &
Education, 60(1), 95–109.
Mayer, R. E. (1997). Multimedia learning: Are we asking the right questions? Educational Psychologist,
32(1), 1–19.
Mayer, R. E. (2003). The promise of multimedia learning: Using the same instructional design methods
across different media. Learning and Instruction, 13(2), 125–139.
Mayer, R. E. (2014a). Cognitive theory of multimedia learning. In R. E. Mayer (Ed.), The Cambridge handbook
of multimedia learning (pp. 43–71). New York, NY: Cambridge University Press.
Mayer, R. E. (2014b). Incorporating motivation into multimedia learning. Learning and Instruction, 29,
171–173.
Nadolny, L. (2017). Interactive print: The design of cognitive tasks in blended augmented reality and print
documents. British Journal of Educational Technology, 48(3), 814–823.
Paas, F., & Sweller, J. (2014). Implications of cognitive load theory for multimedia learning. In R. E. Mayer
(Ed.), The Cambridge handbook of multimedia learning (pp. 27–42). New York, NY: Cambridge University
Press.
Paas, F., & van Merriënboer, J. J. (1994). Instructional control of cognitive load in the training of complex
cognitive tasks. Educational Psychology Review, 6(4), 351–371.
Phillips, L. M., & Norris, S. P. (2009). Bridging the gap between the language of science and the language
of school science through the use of adapted primary literature. Research in Science Education, 39(3),
313–319.
She, H. C., & Chen, Y. Z. (2009). The impact of multimedia effect on science learning: Evidence from eye
movements. Computers & Education, 53(4), 1297–1307.
Shellnut, B., Knowltion, A., & Savage, T. (1999). Applying the ARCS model to the design and development
of computer-based modules for manufacturing engineering courses. Educational Technology Research and
Development, 47(2), 100–110.
Türk, E., & Erçetin, G. (2014). Effects of interactive versus simultaneous display of multimedia glosses on
L2 reading comprehension and incidental vocabulary learning. Computer Assisted Language Learning,
27(1), 1–25.
Wang, J. R., & Chen, S. F. (2014). Exploring mediating effect of metacognitive awareness on comprehension
of science texts through structural equation modeling analysis. Journal of Research in Science Teaching,
51(2), 175–191.
Wang, J. R., Chen, S. F., Fang, I., & Chou, C. T. (2014). Comparison of Taiwanese and Canadian stu-
dents’ metacognitive awareness of science reading, text, and strategies. International Journal of Science
Education, 36(4), 693–713.
Wei, X., Weng, D., Liu, Y., & Wang, Y. (2015). Teaching based on augmented reality for a technical creative
design course. Computers & Education, 81, 221–234.
Yang, F. Y., Chang, C. C., Chen, L. L., & Chen, Y. C. (2016). Exploring learners’ beliefs about science reading
and scientific epistemic beliefs, and their relations with science text understanding. International Journal
of Science Education, 38(10), 1591–1606.
Yilmaz, R. M., Kucuk, S., & Goktas, Y. (2017). Are augmented reality picture books magic or real for pre-
school children aged five to six? British Journal of Educational Technology, 48(3), 824–841.
Yoon, S., Anderson, E., Lin, J., & Elinich, K. (2017). How augmented reality enables conceptual under-
standing of challenging science content. Educational Technology & Society, 20(1), 156–168.