Readdressing the Redundancy Effect: A Cognitive Strategy for E-learning Design

Abstract

This study challenges understandings on the ‘redundancy effect’ of cognitive load theory and visual/verbal classifications of dual-coding theory. Current understandings assert that a multimedia mix of narration and text displayed via e-learning leads to cognitive overload, thus, impeding learning[1][2]. Previous research suggests that for optimal learning to occur, the most effective multimedia mix for e-learning presentation is the use of pictures and narration[3][4][5][6]. The study was undertaken with 90 undergraduate students at a British University. Participants were allocated to one of three groups. Each group used a different multimedia mix of a music e-learning program. The experiment groups were; Redundancy mode, Modality mode and Mixed mode. Participant’s prior level of music learning was assessed, as was their level of knowledge after using the system. Results indicate that the Redundancy mode is most effective for learning, that is, the simultaneous presentation of text and narration enables significantly more optimal learning that modality modes (i.e. Narration and graphics) and Mixed modes (i.e. text and graphics). These findings challenge the currently accepted stance on the redundancy effect in e-learning design.

Full text

1
Journal of Psychological Research | Volume 01 | Issue 02 | July 2019
Distributed under creative commons license 4.0 DOI: https://doi.org/10.30564/jpr.v1i2.580
Journal of Psychological Research
http://ojs.bilpublishing.com/index.php/jpr
ARTICLE
Readdressing the Redundancy Effect: a Cognitive Strategy for
E-learning Design
Sylvie Studente*
Department of AFE, Faculty of Business & Management, Regents University London, Inner Circle, Regents Park, Lon-
don, United Kingdom
ARTICLE INFO ABSTRACT
Article history
Received: 1 March 2019
Accepted: 20 March 2019
Published Online: 31 May 2019
This study challenges understandings on the ‘redundancy effect’ of cog-
nitive load theory and visual/verbal classications of dual-coding theory.
Current understandings assert that a multimedia mix of narration and text
displayed during e-learning leads to cognitive overload, thus, impeding
learning[1,2]. Previous research suggests that for optimal learning to occur,
the most effective multimedia mix for e-learning presentation is the use
of graphics and narration[3-6].
The current study was undertaken with 90 undergraduate students at a
British University. Participants were allocated to one of three groups.
Each group used a different multimedia mix of a music e-learning pro-
gram. Participants received learning material electronically, which in-
volved either a mix of narration and text, graphics and text, or graphics
and narration. Learning was measured by differences in music knowledge
scores obtained before and after receiving the learning material. Results
indicate that the combination of text and narration is most effective for
learning, compared to combinations of graphics and text and graphics and
narration. These ndings challenge the currently accepted stance on the
redundancy effect in e-learning design.
Keywords:
Learning
Memory
Working memory
Graphical user interfaces
*Corresponding Author:
Sylvie Studente,
Department of AFE, Faculty of Business & Management, Regents University London, Inner Circle, Regents Park, London, United
Kingdom;
Email: sylvie.studente@regents.ac.uk.
1. Introduction
Contemporary educational technologies make use
of a range of multimedia elements including text,
graphics, video and sound to present pedagogic
information. However, these elements are often applied
in an ad-hoc manner without considering which mix of
elements will best communicate educational concepts to
students. For example, e-learning platforms may use a
mix of narration and text on screen or just narration with-
out a clear rationale for using multiple or a single presen-
tation mode. Alternatively, an emphasis may be given to
inclusivity in terms of learning styles, such that the more
styles that are addressed the better, without consideration
of what is optimal. For instance, on some virtual learning
platforms subtitles have become a standard feature that
learners have to switch off rather than on. As such, while
technology has made it easier and cheaper to use multiple
presentation modes, educational designers and learners are
not generally informed about the benets and drawbacks
of choosing a particular presentation mode.
Studies in e-learning and psychology have produced

2
Journal of Psychological Research | Volume 01 | Issue 02 | July 2019
Distributed under creative commons license 4.0 DOI: https://doi.org/10.30564/jpr.v1i2.580
mixed and contradictory results about the effectiveness of
using different modes to present learning content. Some
research has supported cognitive load theory, which sug-
gests that conveying excessive information can impede
learning because of the limited capacity of working mem-
ory [7-9] . The outcome of cognitive load has been coined
the redundancy effect, whereby learning is detrimentally
affected by the overloading of either the visual or verbal
processing channels of memory. Other research, in con-
trast, has suggested that learning is more effective when
multiple channels are used [10,11].
Our research attempts to clarify and explain the incon-
sistencies in the literature around the use of multiple pre-
sentation modes (e.g. text, graphics, narration) in e-learn-
ing. In what follows, we show that the inconsistencies in
past research may at be attributed to differences in how
text is treated, as either a verbal rather than visual presen-
tation mode. Moreover, text is a natural form of commu-
nication that is directly encodable and the visual-verbal
categorization may not therefore be representative of how
text is processed. We argue that text paces narrated infor-
mation, and therefore, enhances learning rather than com-
petes for limited working memory resources. It may be
that there is no redundancy effect when text is involved,
as will be shown by our empirical results. Our findings
support and help explain the work conducted by Truman
& Truman[10] and Toh et al[11], who found that e-learning
interfaces that present information via text and narration
simultaneously significantly increase students’ ability to
recall pedagogic concepts.
In what follows, we investigate the effectiveness of
three separate presentation modes on learning and informa-
tion recall: ‘text-only’, ‘text-narration’ and ‘pictorial-nar-
ration’. The e-learning program MOLE (Music Oriented
Learning Environment) is used as a test platform for the
study. We control the effect of several variables that can
interfere with the relationship between presentation modes
and learning including music theory and music instrument
training.
2. Background Motivation
2.1 Human Memory and Dual Coding Theory
Research about the effectiveness of media used in learn-
ing contexts is based on assumptions about the operation
of human memory. There are four well-established mem-
ory processes: control, encoding, storage and retrieval.
Furthermore, the modal model of memory proposed
by Atkinson & Shiffrin[12] identified three sub-stores of
memory: sensory memory, short-term memory and long-
term memory. Information perceived via sensory memory
can be transferred to short term memory via attentional
processes, while information in short-term memory can
be transferred to long-term memory through two primary
conditions: rehearsal of material/information in short-term
memory, and in-depth information processing[1,13]. Unlike
the infinite capacity of long-term memory, short-term
memory is limited in the information it can hold[14].
In addition, the model of working memory proposed by
Baddeley[15] purports that auditory and visual processing
channels are independent, allowing both visual and verbal
representations of information to be held in memory. Two
slave sub-systems are encompassed within working mem-
ory: the articulatory loop and the visuo-spatial sketchpad.
The articulatory loop is responsible for processing and
storing verbal information, whereas the visuo-spatial
sketchpad is responsible for processing and storing visual
information. The ‘central executive’ component of work-
ing memory co-ordinates these sub-systems and allows
referential connections to be formed between visual and
verbal information.
The notion of working memory relates to Paivio &
Csapo’s ‘dual-coding theory’[16]. This theory asserts that
simultaneous multi-channel processing of linguistic infor-
mation is possible whilst providing a symbolic function
to non-verbal objects. This is facilitated by two cognitive
representation units; imagens and logogens[17,18]. Imagens
are concerned with processing pictorial information,
whilst ‘logogens’ are responsible for processing verbal
information. Educational technologies that utilise du-
al-modality are effective for enhancing the recall of peda-
gogic information as they target both the visual and verbal
processing channels. This allows the brain to search along
two ‘paths’ during recall, allowing maximization of an
individual’s response time[19,20]. When translated into the
context of learning, offering learning materials through
the two paths of visual and verbal processing should lead
to more effective learning because information recall is
enhanced.
2.2 Cognitive Load Theory: Redundancy and Mo-
dality Modes
While dual coding theory suggests that using both visual
and verbal presentation of learning material enhances
learning, other theories of memory imply that memory
processes can be overloaded, leading to reduced learning
capacity. More specically, cognitive load theory asserts
that as short term memory is limited in capacity, the use of
repetitive or redundant features of learning material will
overload the cognitive resources of learners, i.e. the visual
or verbal processing channel identied in the dual-coding
theory, culminating in a redundancy effect [7,8,9]. More

3
Journal of Psychological Research | Volume 01 | Issue 02 | July 2019
Distributed under creative commons license 4.0 DOI: https://doi.org/10.30564/jpr.v1i2.580
specically, visual pictures and verbal narration presented
simultaneously with redundant on-screen text increase
cognitive load and can impede learning due to the compe-
tition of resources in working memory[21-24].
In order to avoid the redundancy effect, Clark and
Mayer[3] suggest that learning is most effective when
pedagogic concepts are presented via visual graphics and
verbal narration as opposed to a combination of graphics,
narration and onscreen text[3]. Clark and Mayer’s reason-
ing is that when graphics and words are both presented
together in visual manner, the visual-processing channel
becomes overused. Numerous studies corroborate this
nding [22,25]. In these studies, however, text is considered
to be visual information processed through visual memory
channels as imagens rather than logogens. Rather, as ver-
bal information, text can be processed by logogens via a
verbal-processing channel, alleviating the cognitive over-
load on the visual-processing channel. It is also important
to note that much of the empirical research validating the
redundancy principle has been based on the learning of
scientic concepts and technical material[26,27,24].
2.3 E-learning Design and the Redundancy Effect
Studies within the area of e-learning design have applied
dual-coding theory to the use of pictures and narration
in learning situations owing to the distinction between
visual and verbal entities[4,5,6,21]. Some studies have found
that a mix of pictorial and narration information is more
effective for information recall[2,22,3,4,5], whilst other stud-
ies have found that a mix of text and narration is more
effective[28,29,10,11]. For example, Toh et al[11] investigated
the redundancy effect in multimedia learning via two in-
structional modes: redundant mode and modality mode. In
‘redundant mode’, static pictures and audio narration were
presented with synchronised redundant on-screen text
(verbal overload). In ‘modality mode’, only static pictures
and audio were presented (no overload). Findings revealed
that learners exposed to the redundancy mode achieved
significantly higher comprehension scores than learners
exposed to the modality mode. These ndings suggest that
the redundancy effect does not impede learning; rather,
the use of all of pictures, audio narration, and on-screen
text reduced the cognitive load, and thereby enhanced
learning.
2.4 Challenging Visual-verbal Classications
Research about the redundancy effect in learning assumes
that text and narration are both verbal logogens. Specif-
ically, the simultaneous reading of text whilst listening
to narration are referred to as ‘verbal entities’[15,30,16,18]. In
contrast, graphical images relate to visual memory stores
and processes. Many scholars have adopted the classifi-
cations or ‘visual’ and ‘verbal’ as literal categories in the
design of e-learning. However, in the present study it is
argued that the classication of text as a verbal entity and
images as a visual entity is a false dichotomy. Rather, like
images, the representation of text is visual, as it has a vi-
sual structure.
Furthermore, some scholars have assumed that simul-
taneously presenting text visually and orally can cause
interference between reading and listening to the text
because the speed of reading is usually faster than that
of listening[31,32]. However, we argue here that concurrent
reading and narrated text focuses the learner’s attention on
pacing through the information as opposed to skim read-
ing and thus imparts a deeper level of learning. This view
is supported by Badii & Truman[29] and Truman & Tru-
man[10], who report that processing of visual and auditory
text do not interfere with each other, they are both natural-
ly and directly encodable as forms of communication, and
reinforce rather than impede on learning.
3. Methodology
The purpose of this research is to investigate the redun-
dancy effect in e-learning. In particular, this study will fo-
cus upon the effectiveness of three presentation modes on
pedagogic information recall: narration and text (referred
to as the ‘redundancy’ mode), graphics and narration (re-
ferred to as the modality mode, i.e. visual vs verbal) and
graphics and text (referred to as the mixed visual mode).
3.1 Hypotheses
The following hypothesis was explored:
H1) The redundancy mode will be associated with
greater information recall compared to the modality and
mixed modes.
3.2 Participants and Experimental Procedure
In order to test our hypothesis, an adapted version of
MOLE (Music Oriented Learning Environment) was
used. This software was adapted from the version used
in previous studies by Badii & Truman[29] and Truman
& Truman[10] to consist of three short interactive lessons
relating to music theory fundamentals. MOLE was orig-
inally designed in accordance with the Associated Board
of the Royal School of Music theory guides. The MOLE
software was adapted into three different prototypes to
present multimedia information in accordance with the
conditions under investigation. The prototypes used are
described in Table 1.

4
Journal of Psychological Research | Volume 01 | Issue 02 | July 2019
Distributed under creative commons license 4.0
Table 1. Experimental conditions
Experimental Condition MOLE Prototype Display
Redundancy mode Audio narration and on screen text
Modality mode Static graphics and audio narration
Mixed mode Static graphics and on screen text
Ninety undergraduate students at a British University
(n=90) participated in the study. All participants were ran-
domly selected and equally assigned to one of the condi-
tions shown in Table 1. Participation tool place in a com-
puter lab which accommodated 15 participants at a time.
All participants were provided with a computer running
the MOLE software and a set of headphones. Learning
was captured by participants’ scores on a pre-test and a
post-test, administered prior to and following their session
with MOLE. This was followed uniformly across all con-
ditions.
3.3 Data Collection Protocols
In order to evaluate the actual learning imparted by the
MOLE software, participants completed a paper based
pre-test prior to their learning session and post-tetst im-
mediately after their learning session. The pre-test also
allowed for the assessment of prior knowledge of musical
concepts. Participants were allocated ve minutes to com-
plete the pre-test and fteen minutes to interact with the
MOLE software. Upon completing the learning session
with MOLE, participants were then given ve minutes to
complete the post-test. The post-test included questions
from the pre-test arranged in a random order. The pre-test
and post-test scores were then compared across all three
conditions to ascertain the mode associated with the high-
est recall scores. The total participation time was twenty
ve minutes.
3.4 Ethical Considerations
Participation in this study was voluntary and anonymous.
Participants were assured that they could withdraw their
participation at any time.
4. Results
The responses of participants for each of the 11 questions
of the pre-test and the post-test were averaged, such that
pre-test and post-test scores generated mean. These scores
are presented in Table 2, across each of the modes sepa-
rately.
Table 2. Means and standard deviations of pre-test and
post-test across mode type
Experimental Con-
dition N
Pre-test
Score
M (SD)
Post-Test
Score
M (SD)
Difference
between pre-test
and post-test scores
Redundancy Mode
(Narration and text) 30 4.13 (3.32) 9.53 (1.54) 5.40
Modality Mode
(Static graphics and
narration)
30 2.90 (3.13) 6.13 (2.62) 3.23
Mixed Mode
(Static graphics and
text)
30 1.93 (2.46) 4.70 (2.62) 2.77
As can be seen in Table 2, participants in the redundan-
cy mode attained higher scores in the pre-test and post-
test than those in modality and mixed modes. Those in
the mixed mode (i.e. text and static graphics) attained the
lowest scores on the pre-test and post-test out of all three
conditions. The reason behind the different pre-test scores
across the three conditions can be explained by individual
variables in particular music training, as discussed later in
detail.
A 2 (time: pre-test vs post-test) X 3 (mode type: redun-
dancy vs modality vs mixed) repeated measures ANOVA
was conducted to establish the association between test
performance and interaction with the learning software.
Signicant main effects were observed in that participants
performed significantly better in the post-test compared
to the pre-test, F (1, 87) = 211.75, p = .000. This indicates
that learning was imparted during the e-learning session
across all modes using different combinations of text,
graphics and narration.
In addition, a signicant interaction effect was observed
between mode type and learning performance, F (2, 87)
= 9.65, p = .000. Specically, participants who received
the redundancy mode performed signicantly better in the
post-test, compared to participants who received the mo-
dality mode (p = .001), and participants who received the
mixed mode (p = .000). The participants in the redundan-
cy condition improved their performance on average by
5.40 points against 3.23 points in the modality condition
and 2.77 points in the mixed mode condition. This indi-
cated that the redundancy mode was the most effective
multimedia mix for imparting learning.
5. Additional Analyses
Additional analyses were computed to examine the role
of demographic variables, such as music training, on per-
formance across the pre-test and post-test. A 2 (time: pre-
test vs post-test) X 3 (mode type: redundancy vs modality
vs mixed) X 2 (music theory training: yes vs no) repeated
measures ANOVA was conducted. A significant interac-
DOI: https://doi.org/10.30564/jpr.v1i2.580

5
Journal of Psychological Research | Volume 01 | Issue 02 | July 2019
Distributed under creative commons license 4.0
tion effect was obtained between time and music theory
training, irrespective of the mode type they had received.
Specifically, individuals with music theory training per-
formed signicantly better in the pre-test (M = 4.82, SD =
3.07) and post-test (M = 7.80, SD = 2.80), compared to
individuals with no music theory training (pre-test: M =
1.55, SD = 2.23, post-test: M = 5.95, SD = 3.08), F (1, 83)
= 12.19, p = .000, η2 = .12.
Similarly, a 2 (time: pre-test vs post-test) X 3 (mode
type: redundancy vs modality vs mixed) X 2 (music in-
strument training: yes vs no) repeated measures ANOVA
revealed a signicant interaction effect between time and
music instrument training, irrespective of the mode type
they had received. Specically, individuals with music in-
strument training performed signicantly better in the pre-
test (M = 5.06, SD = 3.05) and post-test (M = 8.28,SD =
2.35), compared to individuals with no music instrument
training (pre-test: M = 1.84, SD = 2.48, post-test: M =
5.96, SD = 3.12), F (1, 84) = 7.60, p = .007, η2= .08. No
signicant differences were identied with regard to gen-
der.
6. Discussion
With regard to our hypothesis, the results from our
study demonstrate that text and concurrent narration leads
to a significantly higher level of learning as opposed to
graphics-text and graphics-narration modes. This nding
opposes views on cognitive load theory, and in particular
the ‘redundancy effect’. That is, rather than impede learn-
ing, a text-and-narration mix is significantly conducive
to learning. There are a number of reasons for this occur-
rence. Firstly, we argue that concurrently read and nar-
rated text focuses a learner’s attention on pacing through
information rather than ‘skim-reading’ and imparting a
deeper level of learning. This concept is substantiated by
Moore[33], who states that “the pace of narration controls
the pace of the material”. In addition, both the written and
spoken word are natural forms of communication, and
thus, directly encodable.
Whereas current understandings on dual-coding theory
suggest that ‘text’ and ‘narration’ are both categorised as
‘verbal entities’[15], we argue that this is a false dichotomy
as the representation of text is a visual display in itself.
Therefore, the simultaneous presentation of text and nar-
ration of identical information within an e-learning system
allows for the simultaneous multi-channel processing of
linguistic information for the learner. Thus, the material
is imparted along two distinct channels in the brain, in-
creasing the learnability of the material. This approach
strengthens the associations of the material being learned,
and is an effective strategy for e-learning design. Our nd-
ings support those reported by Badii and Truman[29] and
Truman and Truman[10] , who reported that text-and-narra-
tion enhance learning performance.
Although all our participants performed better in the
post-test compared to the pre-test, our results indicate that
previous music theory and instrument training led to better
task performance irrespective of the mode through which
the material was delivered. As a result, the results indicat-
ed that the same mix of learning modes, the redundancy
mode, is the most effective one for both novice and expert
learners. Sweller[34] argues that information processing is
likely to differ markedly between novice and expert learn-
ers because expert learners may be more readily able to
process material because of its availability in long-term
memory. However, our results suggest that there is no
difference regarding the effectiveness of learning modes
between students who have different knowledge levels.
Learning modes do not need to be adjusted for novice and
expert learners where information recall is concerned.
However, further research is needed to investigate
how the mix of learning materials interacts with different
knowledge domains and learning types, i.e. recall vs un-
derstanding. As argued by Sweller[35] instructional design
should be adapted to the knowledge domain because the
domain interacts with the capacity and duration of working
memory. Earlier research on cognitive load theory has fo-
cused on examining learning in the domain of scientic and
technical knowledge [1] [32]. For example, Craig and his col-
leagues studied learning related to the process of lightning
formation as a weather condition[1]. In contrast to the earlier
studies pertaining to the learning of scientic concepts, our
study concerned another domain of learning, theory of mu-
sic. Our results are aligned with the ndings of other studies
that have also investigated learning in the area of music (i.e.
Truman & Truman [10] Badii & Truman [29]).
More research is needed to categorise the different
knowledge domains and to examine the significance of
the domain in explaining the effectiveness of learning
materials. Past research has shown that individual qual-
ities inuence learning. For example, it has been argued
that perception defined as the type of information stu-
dents like to receive is the most important dimension of
learning styles[36]. Sensitive students prefer data and
are methodological in their approach, while intuitive
students prefer principles and theories (ibid.). Perception
and other individual qualities may be linked to the
effectiveness of learning modes.
The results of our research have practical implications.
As discussed above, the notion of learning styles has been
criticized in the past because teaching in class rooms has
not been adapted to match the diversity of learning styles
DOI: https://doi.org/10.30564/jpr.v1i2.580

6
Journal of Psychological Research | Volume 01 | Issue 02 | July 2019
Distributed under creative commons license 4.0
displayed by learners. Based on our research, e-learn-
ing material can be easily adapted to deliver knowledge
content through the different modes of narration, text,
and graphics. The overall results suggest that the use of
text and narration together does not impede on learning,
debunking the redundancy effect, and is rather the most
effective for learning.
7. Conclusion
The impact of this research is that the use of text and
narration simultaneously is an effective strategy for
e-learning. This challenges the widely accepted cognitive
load theory, in that the representation of text is visual, and
uses a visual rather than verbal information processing
channel Perhaps for this reason, the use of text and narra-
tion simultaneously does not impede on learning, and is
actually conducive to learning due to the use of two rather
than one learning mode. The use of text and narration to-
gether is also more effective than mixes of graphics-text
and graphics-narration, perhaps due to the greater ease
of encoding communication forms (i.e. text and narra-
tion). Instead, the cognitive load appears to occur where
graphics are concerned. Future research may include de-
lineating the impact of the use of graphics in learning, in
combinations with other forms, as well as investigating
information pacing in concurrent text and narration. This
study has also raised implications for the current classi-
cations of visual and verbal entities of dual coding theory.
References
[1] Craig, S, Gholson, B & Driscoll, D. Animated ped-
agogical agents in multimedia education environ-
ments: effects of agent properties, picture features
and redundancy[J].Journal of Educational Psycholo-
gy, 2002, 94. 428 – 434.
[2] Mayer, R. Applying the science of learning to multi-
media instruction[J].The Psychology of Learning and
Motivation: Cognition in Education, 2011, 55.
[3] Clark, R & Mayer, R. E-learning and the Science of
Instruction (2nd ed.) [M]. 2008.
[4] Kim, S & Lombardino, L. Exploring the Effects of
Narration and Pictures on Learning for Students with
Reading Decits[J]. Clinical Archives of Communi-
cation Disorders, 2017,2(2):116 – 127.
[5] Dunsworth, Q & Atkinson, R. Fostering multimedia
learning of science: Exploring the role of an ani-
mated agent’s image.[J]. Computers and Education,
2007, 49(3): 677 – 690.
[6] Mayer, R. Elements of science of e-learning [J].
Journal of Educational Computing Research. 2003,
29(3): 297 – 313
[7] Clark, R, Nguyen, F, & Sweller, J. Efficiency in
Learning: Evidence-Based Guidelines to Manage
Cognitive Load [M]. 2006.
[8] Sweller, J. Instructional design in technical areas [M].
1999.
[9] Sweller, J.Cognitive load theory [J]. The Psychology
of Learning and Motivation: Cognition in Education,
2011, 55: 38 – 74.
[10] Truman, S & Truman, P. An investigation into the
situated learnability effects of single-and dual-mod-
al systems in education: a report of music-oriented
learning environment and science computer-assisted
studies [J]. British Journal of Educational Technolo-
gy, 2006, 37(11).
[11] Toh, S, Ahmed, W, Munassar, S & Jaafar, W & Yaha-
ya, W. Redundancy effect in multimedia learning: A
closer look[C]. Proceedings of ascilite, 2010.
[12] Atkinson, R & Shiffrin, R. Human memory: a pro-
posed system and its control processes [C]. In K.W
spence and J.T Spence (Eds) The Psychology of
Learning and Motivation: Advances in Research and
Theory, 1968, 2: 89 – 195.
[13] Peterson, L & Peterson, M. Short-term retention of
individual items [J]. Journal of Experimental Psy-
chology, 1959, 53: 193 – 198.
[14] Miller, G. The magical number seven, plus or minus
two; some limits on our capacity for processing in-
formation [J]. Psychological Review. 1956, 63(2): 81
– 97.
[15] Baddeley, A, Working memory [J]. Science, 1992,
255: 556 – 559.
[16] Paivio, A & Csapo, K. Picture superiorty in free re-
call: imagery or dual coding?[J]. Cognitive Psychol-
ogy, 1973, 5: 176 – 206.
[17] Paivio, A. Imagery and Verbal Processes [M]. 1971.
New York; Holt, Rinehart & Winston.
[18] Paivio, A. Dual coding theory and education [C].
Proceedings of the conference on Pathways to Litera-
cy Achievement for High Poverty Children, 2006.
[19] Dix, Finlay, Abowd & Beale. Human Computer In-
teraction [M]. 1993.
[20] Sadoski, M. A dual coding theoretical model of read-
ing [C]. In Alvermann, D. E., Unrau, N. J., Ruddell, R.
B (Eds.) Theoretical Models and Processes of Read-
ing (6th ed.), 2013.
[21] Moreno, R & Mayer, R. Cognitive principles of mul-
timedia learning: The role of modality and contiguity
[J]. Journal of Educational Psychology, 1999, 9(2):
358-368.
[22] Mousavi, S, Low, R, & Sweller, J. Reducing cogni-
tive load by mixing auditory and visual presentation
modes [J]. Journal of Educational Psychology, 1995,
DOI: https://doi.org/10.30564/jpr.v1i2.580

7
Journal of Psychological Research | Volume 01 | Issue 02 | July 2019
Distributed under creative commons license 4.0
87(2): 319–334.
[23] Tindall-Ford, S., Chandler, P., & Sweller, J. When
two sensory modes are better than one [J]. Journal of
Experimental Psychology: Applied, 1997, 3(4): 257-
287.
[24] Larson, M & Lockee, B, Streamlined ID: A Practical
Guide to Instructional Design [M]. 2014.
[25] Ginns, P. Meta-analysis of the modality effect [J].
Learning and Instruction, 2005, 15: 313–331.
[26] Hede, A. Integrated Model of Multimedia Effects on
Learning [J]. Journal of Educational Multimedia and
Hypermedia, 2002, 11(2): 177-191.
[27] Koning, B, Tabbers, H, Rikers, R & Pass, F. Atten-
tion cueing as a means to enhance learning from an
animation [J]. Applied Cognitive Psychology, 2007,
21(6): 731 – 746.
[28] Reed, S. Effect of computer graphics on improving
estimates to algebra word problems [J]. Journal of
Educational Psychology, 1985, 77(3): 285 – 298.
[29] Badii, A & Truman, S. Cognitive factors in interface
design: an e-learning environment for memory per-
formance and retention optimization[C]. Proceedings
of the 8th European Conference on Information
Technology Evaluation , 2001.
[30] Baddeley, A & Hitch, G, Working memory [C]. In G.A
Bower (Ed) Recent Advances in Learning and Moti-
vation, (1974), 8: 47 – 90.
[31] Mandl, H Levin, J. Knowledge Acquisition from Text
and Pictures[M], 1991.
[32] Kalyuga, S., Chandler, P., & Sweller, J. When redun-
dant on-screen text in multimedia technical instruc-
tion can interfere with learning [J]. Human Factors,
2004, 46: 567-581.
[33] Moore, C. Do we really need narration? [N]. 2010:
http://blog.cathy-moore.com/2010/09/do-we-real-
ly-need-narration Accessed: February 2019
[34] Sweller, J. Instructional design consequences of an
analogy between evolution by natural selection and
human cognitive architecture [J]. Instructional Sci-
ence, 2004, 32: 9 – 31.
[35] Sweller, J. Original article: Working memory, long-
term memory, and instructional design [J]. Journal of
Applied Research in Memory and Cognition, 2015.
[36] Felder, R. M. & Silverman, L. K. Learning and
teaching styles in engineering education [J]. Engi-
neering Education, 1988, 78(7): 674.
DOI: https://doi.org/10.30564/jpr.v1i2.580