18 How Multimedia Can Improve

Learning and Instruction

Richard E. Mayer

The Multimedia Principle

Over the course of human history, the primary mode of communication in
education has been with words, including spoken words (e.g., in lectures) and printed
words (e.g., in books). This chapter explores the straightforward idea that human com-
munication can be improved when pictures are added to words. In short, the idea
motivating this chapter is that people learn better from words and pictures than from
words alone. This statement summarizes what has been called the multimedia principle,
which has become a fundamental principle of instructional design based on a growing
body of research evidence (Butcher, 2014; Clark & Mayer, 2016; Mayer, 2009, 2014a).
Multimedia instruction (or a multimedia instructional message) refers to a lesson
containing both words and pictures, where the words can be in spoken form or
printed form and the pictures can be in static form (such as illustrations, charts,
graphs, or photos) or dynamic form (such as animation or video). Multimedia
instruction – educational communications that use words and graphics – can be
presented in books, in live slideshow presentations, in e-learning on computers, or
even in video games or virtual reality.
The rationale for multimedia instruction is both practical and theoretical. On the
practical side, the multimedia principle – i.e., adding pictures to words – has potential to
contribute to the science of instruction by improving how well students understand
academic material; and, on the theoretical side, the multimedia principle has the potential
to contribute to the science of learning by yielding the basis for theories of how people
learn authentic academic content rather than contrived laboratory materials.
The goal of this chapter is to explore the potential of the multimedia principle for
improving how people understand communications about academic content, as mea-
sured by their ability to take what they have learned and apply it to new situations (i.e., to
be able to solve transfer problems). After a brief introduction, this chapter explores the
historical foundations of multimedia learning, the evidence for the multimedia principle,
the theoretical basis for how the multimedia principle works, the instructional implica-
tions of the multimedia principle, and future directions for the multimedia principle.

Preparation of this chapter was supported by Grant N000141262046 from the Office of Naval
Research.

460

https://doi.org/10.1017/9781108235631.019 Published online by Cambridge University Press

 How Multimedia Can Improve Learning and Instruction 461

Historical Overview of the Multimedia Principle

The multimedia principle is at once both an old idea, dating back hundreds
of years to the work of Comenius on the first multimedia textbook in the seventeenth
century, and a new idea, inspired by ever-expanding advances in computer technol-
ogy that allow dazzling graphics in the digital age.
The first children’s multimedia textbook was published in Nuremberg in 1657 by
John Amos Comenius. His book, Orbis Pictus (translated as “the world in pictures”
or simply “visible world”), contained nearly 200 pages, with each page containing a
black line drawing ranging from the parts of a house to a barbershop to creatures that
live on land and water. Each element in the drawing was numbered, and below was
a legend that gave the name in Latin and in the language of the reader for each
numbered object, along with some description. The goal was to provide to children
“a picture and nomenclature of all the chief things in the world . . . so they may see
nothing which they know not how to name and that they may name nothing which
they cannot show.” Comenius’ guiding theory was that “there is nothing in the
understanding, which was not before in the sense,” which is consistent with the
premise of the cognitive theory of multimedia learning that understanding is
enhanced when learners can mentally connect words and graphics. On the practical
side, Orbis Pictus became the bestselling textbook in Europe for a century, insuring
its place as the world’s first educational classic. In the subsequent centuries, picture
books became the staple of children’s textbooks, but modern analyses of the use of
illustrations in textbooks show that most serve little or no pedagogical value and in
some cases can even be distracting (Levin & Mayer, 1993).
Advances in computing technology in today’s digital age have reignited educa-
tors’ interest in multimedia forms of communication because of the ease with which
it is now possible to render illustrations, photos, animation, and video and incorpo-
rate them with audio and text. Access to the Internet, mobile computing, and
interactive virtual reality have made multimedia learning available when and
where the learner wants it. Such advances have prompted calls for expanding the
concept of literacy to new media in which students learn to create and comprehend
multimedia messages (Mayer, 2008). Today’s forms of multimedia instruction have
expanded beyond paper-based formats to live slideshow formats to computer-based
formats, including e-learning, video games, and virtual reality.
However, just because an educational technology exists does not mean that it
will be used productively. For example, Cuban (1986) provides a history of
educational technology in the twentieth century, including motion pictures in the
1920s, radio in the 1930s, educational TV in the 1950s, and machine-based
programmed instruction in the 1960s. In each case, strong claims were made for
the educational potential of the cutting-edge technology of the day, but, within
a decade, it became clear that the technology had failed to revolutionize educa-
tion. Today’s cutting-edge technologies have prompted some visionaries to call
for revolutionizing education – putting multimedia learning experiences online
and making better use of game-like activities to accelerate learning (Gee, 2003;
McGonical, 2011; Prensky, 2006; Schank, 2002).

https://doi.org/10.1017/9781108235631.019 Published online by Cambridge University Press

 462 richard e. mayer

The lessons concerning the educational technologies of the twentieth century

should caution us to replace a technology-centered approach – designing instruction
based on what cutting-edge technology can do without regard to how people learn –
with a learner-centered approach – designing instruction, often multimedia instruc-
tion, based on an understanding of how people learn regardless of the medium used
to deliver the instruction (Mayer, 2009). This chapter takes a learner-centered
approach to designing multimedia instruction in the digital age based on the idea
that instructional media do not cause learning but rather instructional methods cause
learning (Clark, 2001). The next sections examine what the research has to say about
how to design effective multimedia learning experiences for learners.

Evidence Concerning the Multimedia Principle

I stumbled into the field of multimedia learning about thirty years ago as
part of my search for techniques that help people learn in ways that allow them to
subsequently apply what they have learned to new situations. I was trying to figure
out how to help people understand scientific explanations of how cause-and-effect
systems work. For example, consider a verbal description of how a bicycle tire
works: “When the handle is pulled up, the piston moves up, the inlet valve opens, the
outlet valve closes, and air enters the lower part of the cylinder. When the handle is
pushed down, the piston moves down, the outlet valve opens, and air moves out
through the hose.” This is a somewhat accurate – if brief – explanation of how the
pump works, but you may wonder how well people understand this communication.
Our research shows that after students listen to this explanation, they are not able
to generate many useful answers to transfer questions such as the troubleshooting
question, “Suppose you push down and pull up several times but no air comes out.
What could have gone wrong?,” or the redesign question, “What could be done to
make a pump more effective, that is, to move more air more rapidly?” (Mayer &
Anderson, 1991).
However, consider what happens if we add a simple animation depicting the
movement of the handle, piston, and valves in a pump in sync with the narration,
as summarized in Figure 18.1. Our research (Mayer & Anderson, 1991) shows that
students who received multimedia instruction generated more than twice as many
useful answers to transfer questions than students who received only the narration
without any animation. The effect size was greater than d = 1, which means that
adding the animation to the narration pushed transfer performance up by more than
one standard deviation, which is considered a large effect for an instructional
intervention.
Overall, in each of the eight experimental comparisons conducted in our lab
involving brief explanations of how pumps work, how car braking systems work,
how electrical generators work, how lightning storms develop, and how to add
and subtract with signed numbers, students who learned with words and graphics
performed better than students who learned with words alone, yielding a median
effect size of d = 1.39 (Mayer, 2009). Being able to improve transfer

https://doi.org/10.1017/9781108235631.019 Published online by Cambridge University Press

 How Multimedia Can Improve Learning and Instruction 463

“When the handle is pulled up, the piston moves up, the inlet value opens, the outlet valve closes, and air enters the lower part of the cylinder.”

“When the handle is pushed down, the piston moves down, the inlet valve closes, the outlet valve opens, and air moves out through the hose.”

Figure 18.1 Frames from narrated animation on how a bicycle tire pump works

performance by more than a standard deviation is an exciting prospect because

most instructional interventions do not generate that level of effect (Hattie,
2009). The next step in this research program was to determine how best to
design multimedia instruction.

How the Multimedia Principle Works

In order to explain the multimedia principle and to determine how best to
design multimedia instruction, classic theories of learning based mainly on rote
learning of word lists need to be modified and expanded. For example, an explana-
tion for the multimedia principle is provided by the cognitive theory of multimedia
learning, as summarized in Figure 18.2 (Mayer, 2009, 2014a). The cognitive theory
of multimedia learning is based on three key ideas from cognitive science:
Dual-channel principle: The human information processing system contains sepa-
rate channels for verbal and pictorial information (Paivio, 1986; Baddeley,
1992). This is reflected in a verbal channel across the top row of Figure 18.2
and a pictorial channel across the bottom of Figure 18.2.

https://doi.org/10.1017/9781108235631.019 Published online by Cambridge University Press

 464 richard e. mayer

MULTIMEDIA SENSORY LONG-TERM

PRESENTATION MEMORY WORKING MEMORY MEMORY

selecting organizing Verbal

Words Ears Sounds words
words Model
integrating
Prior
Knowledge
selecting organizing Pictorial
Pictures Eyes images Images
images Model

Figure 18.2 Cognitive theory of multimedia learning

Limited capacity principle: Only a few items can be processed in a channel at any
one time (Baddeley, 1992; Sweller, Ayres, & Kalyuga, 2011). This is
reflected in the working memory box in the middle column of Figure 18.2.
Active processing principle: Meaningful learning requires appropriate cognitive
processing during learning, including attending to relevant information,
mentally organizing it into a coherent structure, and integrating it with
relevant prior knowledge (Mayer, 2009; Wittrock, 1989). This is reflected
in the arrows for selecting, organizing, and integrating in Figure 18.2.
The boxes in Figure 18.2 represent memory stores and the arrows represent
cognitive processes during learning. The first box in Figure 18.2 consists of the
multimedia instructional message, which consists of words and pictures. The second
box represents sensory memory – spoken words are held briefly in auditory sensory
memory whereas pictures and printed words are held briefly in visual sensory
memory. If the learner pays attention, as indicated by the selecting arrows, some of
the words and images are transferred to working memory for further processing
within a system that has limited processing capacity in each channel. In working
memory, the learner can arrange words (including printed words transformed from
the visual channel) into a verbal model and images into a pictorial model, as
indicated by the organizing arrows. The final box is long-term memory, which
contains a permanent storehouse of knowledge. The learner activates relevant prior
knowledge and brings it into working memory, where it is connected with the
incoming information and where the verbal and pictorial models are connected, as
indicated by the integrating arrows.
Overall, meaningful learning occurs when the learner engages in appropriate
cognitive processing during learning, including selecting relevant words and images
from the multimedia message for further processing in working memory, mentally
organizing the words into a coherent structure (or verbal model) and the images into
a coherent structure (or pictorial model), and integrating the verbal and pictorial
representations with each other and with relevant prior knowledge activated from
long-term memory. The main challenge in instructional design is to guide learners to
engage in these process, while not overloading their limited processing capacity in
each channel of working memory. This challenge can be addressed by designing
multimedia instruction in ways that minimize extraneous processing (i.e., cognitive

https://doi.org/10.1017/9781108235631.019 Published online by Cambridge University Press

 How Multimedia Can Improve Learning and Instruction 465

processing that does not support the instructional objective, which can be caused by
poor instructional design), manage essential processing (i.e., cognitive processing
aimed at representing the presented material in working memory, which depends on
the complexity of the material for the learner), and foster generative processing (i.e.,
cognitive processing aimed at making sense of the material, which depends on the
learner’s motivation to exert effort). In short, designing effective multimedia instruc-
tion requires not only presenting the relevant material but also guiding the learner’s
cognitive processing of the material.

Implications of the Multimedia Principle for Instructional

Design
In attempting to apply the multimedia principle to practical educational
venues such as classroom instruction, textbooks, and online instruction, it becomes
clear that some ways of incorporating graphics are more effective than others. This
section explores principles for how to design multimedia instruction that are based
on replicated research findings (as reported in primary source publications) and
grounded in cognitive theories of how people learn.
Table 18.1 lists eleven evidence-based principles for the design of multimedia
instruction – including slideshow presentations, textbooks, online instruction, and
educational games. The first column gives the name of the principle, the second
column gives a brief description of the principle, the third column lists the median
effect size based on published experiments comparing the transfer test performance
of students who learned with the standard version of the lesson versus those who
learned with an enhanced version that added the target feature, and the fourth column
shows the number of experiments showing a positive effect out of the total number of
experiments. We focus on principles that yield median effect sizes greater that d =
0.40, which is considered substantial enough to be practically important for educa-
tion (Hattie, 2009).
The first five principles address the instructional goal of reducing extraneous
processing – cognitive processing during learning that does not support the instruc-
tional goal. The theoretical rationale for reducing extraneous processing is that
working memory capacity is limited, so if a learner allocates too much cognitive
processing capacity to extraneous processing during learning there will not be
enough cognitive capacity left to fully engage in essential processing (i.e., cognitive
processing aimed at mentally representing the essential information in working
memory) and generative processing (i.e., cognitive processing aimed at reorganizing
the material and integrating it with relevant knowledge activated from long-term
memory).
The coherence principle is that people learn better when extraneous material is
excluded rather than included (Mayer, 2009; Mayer & Fiorella, 2014). Extraneous
material includes unneeded detail in graphics, background music, or interesting but
irrelevant facts in the text. For example, consider a slideshow lesson on how a virus
causes a cold, such as exemplified in Figure 18.3 (Mayer et al., 2008). In the slide, we

https://doi.org/10.1017/9781108235631.019 Published online by Cambridge University Press

 466 richard e. mayer

Table 18.1 Evidence-based principles for the design of multimedia instruction

Principle Description ES No.
Principles for reducing extraneous processing in multimedia learning
Coherence principle Eliminate extraneous material. 0.86 23/23
Signaling principle Highlight essential material. 0.41 24/28
Spatial contiguity principle Place printed words near corresponding 1.10 22/22
graphics.
Temporal contiguity principle Present corresponding narration and 1.22 9/9
graphics simultaneously.
Redundancy principle Do not add printed onscreen text that 0.86 16/16
duplicates narrated graphics.
Principles for managing essential processing in multimedia learning
Segmenting principle Break lesson into manageable parts. 0.77 10/10
Pretraining principle Provide pretraining in names and 0.75 13/16
characteristics of key elements.
Modality principle Present words in spoken form. 0.76 53/61

Principles for fostering generative processing in multimedia learning

Personalization principle Use conversational language. 0.79 14/17
Voice principle Present spoken text with an appealing 0.74 5/6
human voice.
Embodiment principle Use humanlike gestures. 0.40 13/13

Note. ES = median effect size based on Cohen’s d; No. = number of positive effects out of total number of
comparisons.

have added two sentences at the end of the paragraph that present an interesting but
irrelevant fact (which can be called a seductive detail). Students learned better when
seductive details were excluded from the virus lesson (d = 0.80). Overall, across
twenty-three of twenty-three experimental comparisons, students performed better
on transfer tests when extraneous material was excluded, yielding a median effect
size of d = 0.86, which is considered a large effect. Thus, more learning occurs when
less is presented, that is, when the instructional message is kept as simple as possible.
Some possible boundary conditions are that the coherence principle applies most
strongly for learners with low working memory capacity, when the lesson is pre-
sented at a fast pace not under the learner’s control, and when the extraneous material
is highly distracting (Rey, 2012).
The signaling principle (also called the cueing principle) is that people learn better
when essential material is highlighted (Mayer, 2009; Mayer & Fiorella, 2014; van
Gog, 2014). Highlighting of printed text can involve the use of color, underlining,
bold, italics, font size, font style, or repetition. Highlighting of spoken text can
involve speaking louder or with more emphasis. Highlighting of graphics includes
the use of arrows, color, flashing, and spotlights. For example, in a narrated

https://doi.org/10.1017/9781108235631.019 Published online by Cambridge University Press

 How Multimedia Can Improve Learning and Instruction 467

Step5: Breaking Free from the Host Cell

The new parts are packaged into new virus within the host cell. The new viruses break free from the host
cell. In some cases, they break the host cell open, destroying the host cell in the process, which is called lysis.
In other cases, they punch out of he cell membrane surrounding them, which is called budding. A study
conducted by researchers at Wilkes University in Wilkes-Barre, Pennsylvania, reveals that people who make
love once or twice a week are more immune to colds than folks who abstain from sex. Researchers believe
that the bedroom activity somehow stimulates an immune-boosting antibody called IgA.

STEP5: Breaking free

from the host cell.

Figure 18.3 Do people learn better when we add interesting but extraneous text?

slideshow lesson on how airplanes achieve lift, signaling involved adding headings
such as “Wing Shape: Curved Upper Surface Is Longer,” and emphasizing key
words, such as the emboldened words in the following phrase: “surface on top of
the wing is longer than on the bottom.” Mautone and Mayer (2001) reported better
transfer test performance for students who learned from a signaled multimedia lesson
than from a nonsignaled lesson (d = 0.65). Overall, there was a positive signaling
effect in twenty-four of twenty-eight published experimental comparisons, yielding
a median effect size of d = 0.41, which is considered in the small to medium range.
Some possible boundary conditions are that the signaling effect can be stronger for
low-knowledge learners (Naumann, et al., 2007), when the graphics are complex
(Jeung, Chandler, & Sweller, 1997), and when signaling is used sparingly (Stull &
Mayer, 2007).
The spatial contiguity principle is that people learn better when printed words are
placed near to rather than far from corresponding graphics (Ayers & Sweller, 2014;
Ginns, 2006; Mayer & Fiorella, 2014). For example, Figure 18.4a shows a version of
a lesson on car braking systems with the words presented as a caption at the bottom
of the page or screen (i.e., separated presentation) whereas Figure 18.4b shows the
words placed near the part of the graphic they describe (i.e., integrated presentation).
Johnson and Mayer (2012) reported that students performed substantially better on
transfer tests when they received integrated presentations rather than separated

https://doi.org/10.1017/9781108235631.019 Published online by Cambridge University Press

 468 richard e. mayer

(a) Separated Presentation

(b) Integrated Presentation

Figure 18.4 Which instructional method leads to better learning about braking
systems?

presentations, even though the words and graphics were identical in both treatments
(d = 0.73). Overall, there was a positive effect for spatial contiguity in twenty two out
of twenty two published experiments, yielding a median effect size of d = 1.22,
which is a large effect. Some possible boundary conditions are that the spatial
contiguity effect can be stronger when learners are low in prior knowledge (Mayer
et al., 1995) and when the material is complex (Ayres & Sweller, 2014).
The temporal contiguity principle is that people learn better from a narrated
lesson, when the spoken words are presented simultaneously with the corresponding

https://doi.org/10.1017/9781108235631.019 Published online by Cambridge University Press

 How Multimedia Can Improve Learning and Instruction 469

graphics such as drawings, animation, or video (Ginns, 2006; Mayer & Fiorella,
2014). In successive presentation, the spoken words are presented before (or after)
the graphics are presented. In nine out of nine published experimental comparisons,
students performed better on transfer tests with simultaneous rather than successive
presentations, yielding a median effect size of d = 1.22, which is a large effect. Some
possible boundary conditions are that the temporal contiguity principle is diminished
when the material is very simple (Ginns, 2006), when the material is presented in
very short chunks (Mayer, et al., 1999; Moreno & Mayer, 1999; Schuler et al., 2012),
and when the lesson is slow-paced or under learner control (Michas & Berry, 2000).
The redundancy principle is that people learn better from narration and graphics
than from narration, graphics, and redundant printed text (Adesope & Nesbit, 2012;
Kalyuga & Sweller, 2014; Mayer & Fiorella, 2014). For example, Figure 18.5a
shows a slide from a lesson on lightning that includes animation and narration,
whereas Figure 18.5b shows a slide that includes animation, narration, and onscreen
text that duplicates the narration. Mayer, Heiser, and Lonn (2001) reported that
students performed better on transfer tests when they received a narrated animation
rather than a narrated animation with redundant onscreen text (d = 0.77). Overall, in
sixteen of sixteen published experiments, people performed better on transfer tests
when redundant onscreen text was excluded rather than included, with a median
effect size of d = 0.86, which is a large effect. Some important boundary conditions
are that the redundancy principle may not apply when no graphics are presented
(Moreno & Mayer, 2002), only a few key words are printed on the screen (Mayer &
Johnson, 2008), or the onscreen text is worded differently than the spoken text (Yue,
Bjork, & Bjork, 2013).
The next three principles in Table 18.1 are aimed at managing essential processing
(i.e., cognitive processing for mentally representing the essential material in working

(a) Animation and Narration (b) Animation, Narration, and On-Screen Text

As the air in this updraft cools, water vapor

condenses into water droplets and forms a cloud.

ÒAs the air in this updraft cools, water vapor ÒAs the air in this updraft cools, water vapor
condenses into water droplets and forms a cloudÓ. condenses into water droplets and forms a cloudÓ.

Figure 18.5 Which instructional method leads to better learning from an online
slideshow?

https://doi.org/10.1017/9781108235631.019 Published online by Cambridge University Press

 470 richard e. mayer

memory). When the material is complex for the learner, the amount of essential
processing required to mentally represent the material may overload working mem-
ory capacity. In this case, the learner needs to be able to manage his or her processing
capacity in a way that allows for representing the essential material. Three techni-
ques for accomplishing this goal are breaking the essential material into manageable
parts (i.e., segmenting), learning about the names and characteristics of key elements
before the lesson is presented (i.e., pretraining), and presenting words in spoken form
rather than printed form (i.e., modality).
The segmenting principle calls for breaking a multimedia lesson into manageable
parts (Mayer & Pilegard, 2014). For example, rather than presenting a 2.5 minute
narrated animation on lighting formation as a continuous presentation, suppose we
break it into sixteen segments, each about 10 seconds long with about one sentence,
and allow the learner to click on a CONTINUE key to go to the next segment.
A sample slide is shown in Figure 18.6. This design allows the learner to digest one
step in the process of lightning formation before going on to the next one. Mayer and
Chandler (2001) found that students performed better on transfer tests when they
received segmented rather than continuous lessons on lightning formation, with an
effect size of d = 1.13. The segmenting principle was supported in ten of ten
published experiments, yielding a median effect size of d = 0.79, which is nearly
a large effect. Concerning boundary conditions, the segmenting principle may apply
more strongly for students with low working memory capacity (Lusk et al., 2009)
and for students who are low-achieving (Ayres, 2006).
The pretraining principle calls for teaching students about the names and character-
istics of key elements before presenting the multimedia lesson (Mayer & Pilegard,
2014). For example, before presenting a narrated animation depicting how a car’s
braking system works, students can be presented with a diagram of the braking system
showing the key parts – e.g., brake petal, piston, wheel cylinders, and brake shoes – as

Figure 18.6 Do people learn better when a CONTINUE button is added after
each segment?

https://doi.org/10.1017/9781108235631.019 Published online by Cambridge University Press

 How Multimedia Can Improve Learning and Instruction 471

shown in Figure 18.7. When the learner clicks on a part, such as the piston, the
computer shows that the part is called a piston and tells the learner that the piston can
move forward and back. Mayer, Mathias, and Wetzell (2002) found that students who
received this pretraining before the multimedia lesson performed better on transfer

Figure 18.7 Do people learn better when they receive pretraining in the names
and characteristics of the key elements?

https://doi.org/10.1017/9781108235631.019 Published online by Cambridge University Press

 472 richard e. mayer

tests than those who received no pretraining (d = 0.86). In thirteen of sixteen published
experiments, pretrained learners performed better on transfer tests than non–pretrained
learners, with a median effect size of d = 0.75, which is in the medium range.
An important boundary condition is that the pretraining principle may apply to low-
knowledge but not high-knowledge learners (Pollock, Chandler, & Sweller, 2002).
The modality principle is that people learn better from multimedia presentations
when the words are spoken rather than printed (Low & Sweller, 2014; Mayer &
Pilegard, 2014). The rationale is that the visual channel may become overloaded by
having to process both graphics and printed words, but processing capacity in the
visual channel can be freed up when the words are spoken and therefore processed in
the verbal channel. For example, Figure 18.8(a) shows a frame from a narrated
animation on lightning whereas Figure 18.8(b) shows a frame from the same lesson
with words printed on the screen as a caption. Mayer and Moreno (1998) found strong
evidence that students performed better on transfer tests when the words were spoken
rather than printed for this fast-paced animation that was presented under system
control (d = 1.49). The modality principle is the most studied of all the multimedia
design principles, with positive effects found in fifty-three of sixty-one published
experiments, yielding a median effect size of d = 0.76, which is in the medium range.
Some of the boundary conditions identified in the literature are that the effect can be
eliminated when the lesson is self-paced (Tabbers, Martens, & van Merrienboer, 2004)
or when the verbal segments are long and complex for learners (Schuler et al., 2012).
The final three principles in Table 18.1 are intended to foster generative proces-
sing, that is, cognitive processing aimed at making sense of the presented material.
Even if cognitive capacity is available, learners may not be motivated to use it to
process the material deeply. Social cues can help motivate learners to engage in
deeper processing because people tend to want to understand what a communication
partner is telling them. Thus, principles based on social cues are intended to make
learners feel as if they are in a conversation with the instructor, that is, they feel that

(a) Animation and Narration (b) Animation, Narration, and On-Screen Text

As the air in this updraft cools, water vapor

condenses into water droplets and forms a cloud.
ÒAs the air in this updraft cools, water vapor ÒAs the air in this updraft cools, water vapor
condenses into water droplets and forms a cloudÓ. condenses into water droplets and forms a cloudÓ.

Figure 18.8 Which instructional method leads to better learning from an online
slideshow?

https://doi.org/10.1017/9781108235631.019 Published online by Cambridge University Press

 How Multimedia Can Improve Learning and Instruction 473

Table 18.2 Portions of nonpersonalized and personalized text from a narrated animation on how
the human respiratory system works
Nonpersonalized Version
“During inhaling, the diaphragm moves down creating more space for the lungs, air enters through the
nose or mouth, moves down through the throat and bronchial tubes to tiny air sacs in the lungs . . . ”
Personalized Version
“During inhaling, your diaphragm moves down creating more space for your lungs, air enters through
your nose or mouth, moves down through your throat and bronchial tubes to tiny air sacs in your
lungs . . . ”

the instructor is a social partner. This approach yields the newest of the multimedia
design principles, including using conversational language (personalization princi-
ple), using an appealing human voice (voice principle), and using humanlike ges-
tures (embodiment principle).
The personalization principle is that people learn better from a multimedia lesson
when the words are in conversation style rather than formal style (Ginns, Martin, &
Marsh, 2013; Mayer, 2014b). For example, Table 18.2 shows a portion of the words
from a lesson on how the human respiratory system works presented in third-person
form (e.g., “the lungs”) or in first- and second-person form (e.g., “your lungs”).
Students performed better on a transfer test when the words were in conversational
style (i.e., in first- and second-person form), with an effect size of d = 0.79 (Mayer
et al., 2004). Overall, there were positive effects in fourteen of seventeen published
experiments on personalization (including polite vs. direct wording), yielding
a median effect size of d = 0.79 which is nearly a large effect. Concerning boundary
conditions, the personalization principle works best for less knowledgeable learners
(McLaren, DeLeeuw, & Mayer, 2011a, 2011b; Wang et al., 2008) and lower achiev-
ing learners (Yeung et al., 2009) as well as with shorter lessons (Ginns et al., 2013).
The voice principle is that people learn better from multimedia lessons involving
spoken words when the narrator has an appealing human voice rather than a machine
voice or an unappealing voice (Mayer, 2014b). In five out of six experimental
comparisons, people learned better from narrated animations – such as a 2.5 minute
animated presentation on lightning formation (Mayer, Sobko, & Mautone, 2003) –
when the words were spoken in an appealing human voice rather than in a machine
voice or in an unappealing human voice, yielding a median effect size of d = 0.74.
An important boundary condition is that the positive impact of a human voice can be
overturned by the use of negative social cues such as presenting an onscreen agent
that does not engage in humanlike gesturing (Mayer & DaPra, 2012).
The embodiment principle is that people learn better from multimedia lessons in
which an onscreen agent or instructor uses humanlike gesture (Mayer, 2014b). For
example, Mayer and DaPra (2012) presented students with a narrated slideshow
lesson on how solar cells work in which an onscreen animated pedagogical agent
stood next to the slide (as shown in Figure 18.9) and either displayed humanlike
gestures or did not move during the lesson. Students learned better when the

https://doi.org/10.1017/9781108235631.019 Published online by Cambridge University Press

 474 richard e. mayer

Figure 18.9 Do people learn better when an onscreen agent uses humanlike
gestures or stands still?

onscreen agent used humanlike gestures (d = 0.92). Overall, in thirteen of thirteen

published experiments, students performed better on transfer tests when they learned
from onscreen agents or instructors that exhibited humanlike gestures, yielding
a median effect size of d = 0.40. Concerning boundary conditions, the embodiment
effect is reduced or eliminated when the lesson contains negative social cues such as
a machine voice (Mayer & DaPra, 2012).
What happens when we combine these principles within the context of an actual
classroom? Issa and colleagues (2013) compared how beginning medical students
learned from a standard slideshow lesson or from a lesson in which the slides were
modified based on multimedia design principles such as in Table 18.1. On a transfer
test administered four weeks later, students in the modified group outperformed
those in the standard group with an effect size of d = 1.17, even though the content
was the same. This study – and similar ones (Harskamp, Mayer, & Suhre, 2007; Issa
et al., 2011) – suggest that applying multimedia principles to the design of classroom
instruction can greatly increase student learning.
What happens when we apply multimedia design principles to the design of
educational games? In a more recent review, Mayer (2014c) found that students
learned more from educational games when they were based on the personalization,
pretraining, modality, and redundancy principles. Thus, principles of multimedia

https://doi.org/10.1017/9781108235631.019 Published online by Cambridge University Press

 How Multimedia Can Improve Learning and Instruction 475

design appear to apply in game-based venues as well as in paper-based and compu-

ter-based lessons.
In summary, Table 18.1 lists eleven evidence-based principles of multimedia
design that are intended to maximize the effectiveness of multimedia instruction.
Each has important boundary conditions, largely consistent with the cognitive theory
of multimedia learning.

Future Directions for the Multimedia Principle

Some potential future directions for principles of multimedia learning are
listed below.
1. Studies examining design principles for multimedia instruction have been forged
mainly in short-term laboratory studies, so future research is needed to examine
how the principles apply in more authentic educational environments such as with
real students in real classrooms and over longer time periods including the use of
delayed tests.
2. Studies examining design principles for multimedia instruction have been conducted
mainly looking at one feature at a time, so future research is needed to determine
what happens when multiple features are used in conjunction with each other.
3. Studies on design principles for multimedia instruction have focused mainly on
helping students learn conceptual knowledge (such as explanations of how
a scientific system works), so future research is needed to determine whether
the principles also apply for other kinds of learning objectives in the revised
Bloom’s taxonomy including learning facts, procedures, and strategies
(Anderson, et al., 2001).
4. As the field progresses, it is useful to develop converging evidence on the
boundary conditions for each multimedia design principle. In particular, most
of the research supporting this chapter was conducted with low-knowledge
learners (or beginners), and there is emerging evidence that the principles may
not apply to high-knowledge learners (or experts). Kalyuga (2014) uses the term
expertise reversal effect to refer to the finding that instructional manipulations
that benefit learning for beginners do not work for experts or may even be
detrimental to experts. Research is needed to determine if the expertise reversal
principle applies to each multimedia design principle.
5. In light of technological advances, it is useful to determine the extent to which
multimedia design principles apply to new media such as learning in virtual
reality, video games, massive open online courses (MOOCs), and interactive
e-books.
6. Finally, as the field progresses, the theories underlying the principles need to be
sharpened and expanded to include motivational, metacognitive, affective, and
social factors.
What should not change in the future is a focus on rigorous scientific methods
grounded in research-based theories of how people learn. Over the past thirty years,

https://doi.org/10.1017/9781108235631.019 Published online by Cambridge University Press

 476 richard e. mayer

educational and cognitive psychology have amassed encouraging evidence that

human understanding can be improved substantially when we add appropriate
graphics to text. The power of multimedia learning has useful practical implications
for the design of instruction and useful theoretical implications for the science of
learning. In multimedia learning, pictures do not replace words but rather work
together with words to form an instructional message that results in deeper
understanding.

References
Adesope, O. O. & Nesbit, J. C. (2012). Verbal redundancy in multimedia learning environ-
ments: A meta-analysis. Journal of Educational Psychology, 104, 250–263.
Anderson, L. W., Karthwohl, D. R., & Airasian, P. W. et al. (2001). A taxonomy for learning,
teaching, and assessing: A revision of Bloom’s taxonomy of educational objectives.
New York: Longman.
Ayres, P. (2006). Impact of reducing intrinsic cognitive load on learning in a mathematical
domain. Applied Cognitive Psychology, 20(3), 287–298.
Ayres, P. & Sweller, J. (2014). The split attention principle in multimedia learning.
In R. E. Mayer (ed.), The Cambridge handbook of multimedia learning, 2nd edn
(pp. 206–226). New York: Cambridge University Press.
Baddeley, A. D. (1992). Working memory. Science, 255, 556–559.
Butcher, K. R. (2014). The multimedia principle. In R. E. Mayer (ed.), The Cambridge
handbook of multimedia learning, 2nd edn (pp. 174–205). New York: Cambridge
University Press.
Clark, R. C. & Mayer, R. E. (2016). e-Learning and the science of instruction, 4th edn.
Hoboken, NJ: Wiley.
Clark, R. E. (2001). Learning from media. Greenwich, CT: Information Age Publishing.
Comenius, J. A. (1887). Orbis pictus. Syracuse, NY: Bardeen.
Cuban, L. (1986). Teachers and machines: The classroom use of technology since 1920.
New York: Teachers College Press.
Gee, J. P. (2003). What video games have to teach us about learning and literacy. New York:
Palgrave Macmillan.
Ginns, P. (2006). Integrating information: A meta-analysis of spatial contiguity and temporal
contiguity effects. Learning and Instruction, 16, 511–525.
Ginns, P., Marin, A. J., & Marsh, H. M. (2013). Designing instructional text for conversational
style: A meta-analysis. Educational Psychology Review, 25, 445–472.
Harskamp, E. G., Mayer, R. E., & Suhre, C. (2007). Does the modality principle for multi-
media learning apply to science classrooms? Learning and Instruction, 17,
465–477.
Hattie, J. (2009). Visible learning. New York: Routledge.
Issa, N., Mayer, R. E., Schuller, S., Wang. E., Shapiro, M. B., & DaRosa, D. A. (2013).
Teaching for understanding in medical classrooms using multimedia design
principles. Medical Education, 47, 388–396.
Issa, N., Schuller, M., Santacaterina, S., Shapiro, M., Wang, M., Mayer, R. E., &
DaRosa, D. A. (2011). Applying multimedia design principles enhances learning
in medical education. Medical Education, 45, 818–826.

https://doi.org/10.1017/9781108235631.019 Published online by Cambridge University Press

 How Multimedia Can Improve Learning and Instruction 477

Jeung, H., Chandler, P., & Sweller, J. (1997). The role of visual indicators in dual sensory
mode instruction. Educational Psychology, 17, 329–433.
Johnson, C. & Mayer, R. E. (2012). An eye movement analysis of the spatial contiguity effect
in multimedia learning. Journal of Experimental Psychology: Applied, 18, 178–191.
Kalyuga, S. (2014). The expertise reversal principle in multimedia learning. In R. E. Mayer
(ed.), The Cambridge handbook of multimedia learning, 2nd edn (pp. 576–597).
New York: Cambridge University Press.
Kalyuga, S. & Sweller, J. (2014). The redundancy principle in multimedia learning.
In R. E. Mayer (ed.), The Cambridge handbook of multimedia learning, 2nd edn
(pp. 247–262). New York: Cambridge University Press.
Levin, J. R. & Mayer, R. E. (1993). Understanding illustrations in text. In B. K. Britton,
A. Woodworth, & M. Binkley (eds.), Learning from textbooks (pp. 95–113).
Hillsdale, NJ: Lawrence Erlbaum.
Low, R. & Sweller, J. (2014). The modality principle in multimedia learning. In R. E. Mayer
(ed.), The Cambridge handbook of multimedia learning, 2nd edn (pp. 227–246).
New York: Cambridge University Press.
Lusk, D. L., Evans, A. D., Jeffrey, T. R., Palmer, K. R., Wikstrom, C. S., & Doolittle, P. E.
(2009). Multimedia learning and individual differences: Mediating the effects of
working memory capacity with segmentation. British Journal of Educational
Technology, 40(4), 636–651.
Mautone, P. D. & Mayer, R. E. (2001). Signaling as a cognitive guide in multimedia learning.
Journal of Educational Psychology, 93, 377–389.
Mayer, R. E. (2008). Multimedia literacy. In D. J. Leu, J. Coiro, M. Knobel, & C. Lankshear
(eds.), Handbook of research on new literacies (pp. 359–377). Mahwah, NJ:
Lawrence Erlbaum.
(2009). Multimedia learning, 2nd edn. New York: Cambridge University Press.
(ed.) (2014a). The Cambridge handbook of multimedia learning, 2nd edn. New York:
Cambridge University Press.
(2014b). Principles based on social cues in multimedia learning: Personalization, voice,
image, and embodiment principles. In R. E. Mayer (ed.), The Cambridge handbook
of multimedia learning, 2nd edn (pp. 345–368). New York: Cambridge University
Press.
(2014c). Computer games for learning: An evidence-based approach. Cambridge, MA:
MIT Press.
Mayer, R. E. & Anderson, R. B. (1991). Animations need narrations: An experimental test of a
dual-coding hypothesis. Journal of Educational Psychology, 83, 484–490.
Mayer, R. E. & Chandler, P. (2001). When learning is just a click away: Does simple user
interaction foster deeper understanding of multimedia messages? Journal of
Educational Psychology, 93, 390–397.
Mayer, R. E. & DaPra, C. S. (2012). An embodiment effect in computer-based learning with
animated pedagogical agent. Journal of Experimental Psychology: Applied, 18,
239–252.
Mayer, R. E., Fennell, S., Farmer, L., & Campbell, J. (2004). A personalization effect in
multimedia learning: Students learn better when words are in conversational style
rather than formal style. Journal of Educational Psychology, 96, 389–395.
Mayer, R. E. & Fiorella, L. (2014). Principles for reducing extraneous processing in multi-
media learning: Coherence, signaling, redundancy, spatial contiguity, and temporal

https://doi.org/10.1017/9781108235631.019 Published online by Cambridge University Press

 478 richard e. mayer

contiguity. In R. E. Mayer (ed.), The Cambridge handbook of multimedia learning,

2nd edn (pp. 345–368). New York: Cambridge University Press.
Mayer, R. E., Griffith, E., Jurkowitz, I. T. N., & Rothman, D. (2008). Increased interestingness
of extraneous details in multimedia science presentation leads to decreased learning.
Journal of Experimental Psychology: Applied, 14, 329–339.
Mayer, R. E., Heiser, H., & Lonn, S. (2001). Cognitive constraints on multimedia learning:
When presenting more material results in less understanding. Journal of
Educational Psychology, 93, 187–198.
Mayer, R. E. & Johnson, C. I. (2008). Revising the redundancy principle in multimedia
learning. Journal of Educational Psychology, 100, 380–386.
Mayer, R. E., Mathias, A., & Wetzell, K. (2002). Fostering understanding of multimedia
messages through pre-training: Evidence for a two-stage theory of mental model
construction. Journal of Experimental Psychology: Applied, 8, 147–154.
Mayer, R. E. & Moreno, R. (1998). A split-attention effect in multimedia learning: Evidence
for dual processing systems in working memory. Journal of Educational
Psychology, 90, 312–320.
Mayer, R. E., Moreno, R., Boire, M., & Vagge, S. (1999). Maximizing constructivist learning
from multimedia communications by minimizing cognitive load. Journal of
Educational Psychology, 91, 638–643.
Mayer, R. E. & Pilegard, C. (2014). Principles for managing essential processing in multi-
media learning: Segmenting, pretraining, and modality principles. In R. E. Mayer
(ed.), The Cambridge handbook of multimedia learning, 2nd edn (pp. 379–315).
New York: Cambridge University Press.
Mayer, R. E., Sobko, K., & Mautone, P. D. (2003). Social cues in multimedia learning: Role of
speaker’s voice. Journal of Educational Psychology, 95, 419–425.
Mayer, R. E., Steinhoff, K., Bower, G., & Mars, R. (1995). A generative theory of textbook
design: Using annotated illustrations to foster meaningful learning of science text.
Educational Technology Research and Development, 43, 31–43.
McGonical, J. (2011). Reality is broken: Why games make us better and how they can change
the world. New York: Penguin.
McLaren, B. M., DeLeeuw, K. E., & Mayer, R. E. (2011a). Polite web-based intelligent tutors:
Can they improve learning in classrooms? Computers and Education, 56, 574–584.
(2011b). A politeness effect in learning with web-based intelligent tutors. International
Journal of Human-Computer Studies, 69, 70–79.
Michas, I. C. & Berry, D. (2000). Learning a procedural task: Effectiveness of multimedia
presentations. Applied Cognitive Psychology, 14, 555–575.
Moreno, R. & Mayer, R. E. (1999). Cognitive principles of multimedia learning: The role of
modality and contiguity. Journal of Educational Psychology, 91, 358–368.
(2002). Verbal redundancy in multimedia learning: When reading helps listening. Journal
of Educational Psychology, 94, 156–163.
Naumann, J., Richter, T., Flender, J., Cristmann, U., & Groeben, N. (2007). Signaling in
expository hypertexts compensates for deficits in reading skill. Journal of
Educational Psychology, 99, 791–807.
Paivio, A, (1986). Mental representations: A dual-coding approach. Oxford: Oxford
University Press.
Pollock, E., Chandler, P., & Sweller, J. (2002). Assimilating complex information. Learning
and Instruction, 12, 61–86.
Prensky, M. (2006). Don’t bother me mom – I’m learning! St. Paul, MN: Paragon House.

https://doi.org/10.1017/9781108235631.019 Published online by Cambridge University Press

 How Multimedia Can Improve Learning and Instruction 479

Rey, G. D. (2012). A review and meta-analysis of the seductive detail effect. Educational
Psychology Review, 7, 216–237.
Schank, R. C. (2002). Designing world-class e-learning. New York: McGraw-Hill.
Schuler, A., Scheiter, K., Rummer, R., & Gerjets, P. (2012). Explaining the modality effect in
multimedia learning: Is it due to a lack of temporal contiguity with written text and
pictures? Learning and Instruction, 22, 92–102.
Stull, A. & Mayer, R. E. (2007). Learning by doing versus learning by viewing: Three
experimental comparisons of learner-generated versus author-provided graphic
organizers. Journal of Educational Psychology, 99, 808–820.
Sweller, J., Ayres, P., & Kalyuga, S. (2011). Cognitive load theory. New York: Springer.
Tabbers, H. K., Martens, R. L., & van Merrienboer, J. J. G. (2004). Multimedia instructions
and cognitive load theory: Effects of modality and cueing. British Journal of
Educational Psychology, 74, 71–81.
van Gog, T. (2014). The signaling (or cueing) principle in multimedia learning. In R. E. Mayer
(ed.), The Cambridge handbook of multimedia learning, 2nd edn (pp. 263–278).
New York: Cambridge University Press.
Wang, N., Johnson, W. L., Mayer, R. E., Rizzo, P., Shaw, E., & Collins, H. (2008).
The politeness effect: Pedagogical agents and learning outcomes. International
Journal of Human-Computer Studies, 66, 98–112.
Wittrock, M. C. (1989). Generative processes of comprehension. Educational Psychologist,
24, 345–376.
Yeung, A., Schmid, S., George, A. V., & King, M. M. (2009). Using the personalization
hypothesis to design e-learning environments. In M. Gupta-Bhowon, S. Jhaumer-
Laulloo, H. K. L. Wah, & P. Ramasami (eds.), Chemistry education is the ICT age
(pp. 287–300). Berlin: Springer.
Yue, C. L., Bjork, E. L., & Bjork, R. A. (2013). Reducing verbal redundancy in multimedia
learning: An undesired desirable difficulty? Journal of Educational Psychology,
105, 266–277.

https://doi.org/10.1017/9781108235631.019 Published online by Cambridge University Press