It's a common misconception that the more you give learners the better. Figure 5.1 shows a typical example. On this e-learning screen from our overloaded Excel lesson on the CD, we have explained the spreadsheet example with text on the screen and also with audio narration of that same text. The rationale for delivering content in dual modes is that they offer twice the learning opportunity. In addition, many believe that two modes accommodate learners who have visual learning styles as well as learners who have auditory learning styles. Another related misconception is that we can improve motivation and learning by adding interesting information to our basic instructional materials. Unrelated themes and games are commonly added to e-learning as a countermeasure for high dropout rates and to accommodate younger learners raised on computer games. An example of irrelevant information is found in Figure 5.1 in the "Did You Know" box in the lower left corner.
In this chapter we review evidence that recommends a lean approach to your training materials by eliminating extraneous content in the form of pictures, words, or audio that may be related to the main topic but is not relevant to the instructional goal. We also recommend eliminating multiple modes of expression that communicate the same information, such as the text and audio shown in Figure 5.1. Mayer and Moreno (2003) recommend weeding as a solution to over-inflated instruction. Alternatively, you can avoid weeding by simply avoiding extraneous content or modes of expression in the original course design.
By paring your content and delivery modes down to the essentials, you not only make learning more efficient, but you save development time as well. No need to spend extra resources on those entertaining but extraneous motivational stories or themes. No need to invest in redundant displays of content when one will do the job better. We know that working memory has a very limited capacity and we need to reserve those resources for learning. The most effective approach is to simply present the learner with the minimum required to achieve the instructional goals.
Guideline 8 recommends two ways to limit your words to the minimum needed to present the content related to the learning goal: (1) write concise instructional materials and (2) eliminate or resequence technical details.
Vigorous writing is concise (Strunk & White, 2000). Technical writers have long known that succinct rendering of content is more effective than verbose versions. In the 1980s, John Carroll (1992), as a result of many usability tests, designed and recommended computer manuals that he termed minimalist. In developing effective computer reference materials, he recommends that "One must deliberately not explain everything, only what must be explained. . . . Perhaps most important, when there is a problem one must not automatically add functions or build additional training modules; one must consider removing the function or documentation material associated with the problem" (p. 344).
In concise writing every word contributes to the sentence and every sentence contributes to the learning goal. For example, Figures 5.2 and 5.3 show a flabby and a lean text version from our Excel load managed e-lesson on the CD. As you can see, our lean screen includes thirty fewer words than the inflated example. By ensuring that all your materials are as concise as possible, you can save instructional time and get better learning results!
Figure 5.3. A Concise Version of the Text in Figure 5.2 from Our Excel Load Managed Lesson on Our CD.
WHAT THE RESEARCH SAYS ABOUT CONCISE MATERIALS
Mayer, Bove, Bryman, Mars, and Tapangco (1995) conducted three experiments that compared learning about weather processes from summary lesson versions with learning from a complete passage. The summary version consisted of five captioned illustrations, two of which are shown in Figure 5.4. In Experiment 1, college students studied one of three lesson versions for five minutes. The versions were (1) a summary with forty-eight words and five illustrations; (2) a full passage with five hundred words and no illustrations; and (3) a combination of the full passage and the summary. The full passage alone (Version 2) resulted in significantly less learning than either the summary alone (Version 1) or the full passage plus summary (Version 3), which were equivalent. As we discussed in Chapter 3, learning is enhanced when words are illustrated with complementary explanatory visuals.
In another experiment, the benefits of brevity were tested by comparing summaries in which the same illustrations were described with varying numbers of words. The lean version (which was the same as that used in the previous experiment) described the illustrations with captions of fifty words, the middle version added one hundred descriptive words, and the verbose version added 550 words. As you can see in Figure 5.5, the lean version led to the most learning.
The research team concluded that (1) a summary that includes a combination of relevant words and pictures is better than a summary that includes either words or pictures alone and (2) a summary with words and pictures is more effective when it contains a small rather than a large amount of text. To understand the weather process, both words and pictures were needed since neither one could stand on its own. The research team cautioned, however, that "It would be incorrect to conclude that in all cases students can learn as much from studying a summary as from studying a full lesson along with a summary" (p. 72).
The best approach is to provide just enough content, that is, words and pictures, needed to help the learner build the desired schemas. While in the weather lesson, only a few pictures and words were sufficient, other instructional goals might require more.
An effective summary is one which is concise, coherent, and coordinated (Mayer, Bove, Bryman, Mars, & Tapangco, 1995). That is, it uses fewer words to present the content, the words and pictures relate to each other, and, in combination, the words and pictures are directly relevant to the instructional goal. In the above research, all learners were given the same amount of study time. However, if learners had been allowed to study until they understood the material, it is highly likely that the concise versions would have required less study time than the more verbose. Therefore more concise text not only leads to better learning but would save instructional time as well. Since the most expensive element of any training program is the time learners spend away from the job, saving instructional time will directly contribute to cost benefit.
It's a common tendency of subject-matter experts to want to tell it all. As experts, they are not subject to the same cognitive load that novices will experience. Therefore, often they don't even realize they have overloaded their instruction with way too much content for the audience and for the instructional goal. Good instructional design sorts out the need-to-know from the nice-to-know and tosses out the nice-to-know. Here we look at research that shows that omitting technical content that is not required to understand the lesson improves learning.
WHAT THE RESEARCH SAYS ABOUT OMITTING EXTRANEOUS TECHNICAL DETAILS
Mayer and Jackson (2004) prepared concise and expanded versions of a lesson on wave formation for delivery in both paper and multimedia formats. The concise materials used 553 words and five illustrations, whereas the expanded version included 980 words and eleven illustrations. The expanded version incorporated mathematical equations, symbols, and numerical computations relevant to three processes of ocean waves. College students were given seven minutes to study either the concise or expanded version. Following the study period, they took a test that required them to apply what they learned to answer conceptual questions such as: "Why do ocean waves move toward the shore?"
In a second experiment using the paper-based materials, learners were allowed to study the materials as long as they wished beyond a minimum time period. As you can see in Figure 5.6, all concise lesson versions that omitted quantitative details, whether presented on paper or via computer animations, resulted in significantly better learning. The authors suggested that, when learning a cause-and-effect process, learners first form a qualitative mental model. Therefore they recommend that detailed quantitative information be deleted. Should the instructional goal require mathematical computations, sequence them after learners have built an initial qualitative understanding.
Under Guideline 8, we recommended writing concisely. We also recommended omitting technical content not needed to achieve the instructional objective. Now we turn to extraneous content that is used primarily to add interest to lessons. Under Guideline 9 we review evidence recommending omission of words, pictures, and audio added primarily for the purpose of interest or appeal. While well intended, such adjuncts run the risk of depressing learning.
e-Learning has been widely acknowledged for its high rate of dropout. Some suggest that younger learners weaned on computer games are especially prone to drop out from multimedia instruction that is devoid of the motivational elements contained in such games. As a countermeasure, some instructional programs have adopted an approach called edutainment. In edutainment adjunct graphics, music, themes, vignettes, and games designed to make the instruction more appealing are added to the basic instructional materials. Some adjuncts are completely unrelated to the content, such as embedding technical computer training in a "Tomb Raiders" game theme augmented with high end graphics and music. Other adjuncts are related to the content but unnecessary to the instructional goal. For example, the box in the lower left corner of the screen in Figure 5.1 is an adjunct that we created to add interest to our Excel lesson.
In keeping with our chapter theme of less is more, you won't be surprised that evidence shows that adding motivational content—even content topically related to the lesson—depresses learning. The good news is that you can save the resources required to create these additions and the learner's time invested in interacting with them. Instead of edutainment, you can invest your resources in promoting what instructional psychologists call cognitive motivation.
Related or unrelated themes, vignettes, or games added to technical materials in order to increase appeal are examples of emotional sources of motivation. These kinds of high-appeal additions are based on the theory that increasing the emotional impact of technical instruction by adding humor or interest leads to increased student engagement with the lesson and subsequently greater learning. In contrast, cognitive sources of motivation include instructional methods added to support the basic learning processes summarized in Chapter 2. Some examples include graphic organizers placed at the start of a lesson to preview the relationships among the content, an effective worked example to aid learners in building a mental model, as well as the many instructional methods we recommend throughout this book to manage cognitive load. Research we review next suggests that you omit additions designed to increase emotional motivation and invest your resources in cognitive motivational elements instead.
WHAT THE RESEARCH SAYS ABOUT ADDING HIGH EMOTIONAL INTEREST MATERIAL
Harp and Mayer (1997) compared two versions of a short lesson on weather. The basic version contained 550 words and five captioned illustrations similar to the ones shown in Figure 5.4. Three high interest versions were created by adding 150 words, five pictures, or 150 words plus five pictures that described related but irrelevant facts about lightning. For example, one passage discussed the number of Americans killed by lightning and the fact that swimmers are highly vulnerable to lightning strikes. Another passage described what happens to airplanes that are struck by lightning during flight. Yet another passage discussed a football player who was injured by lightning. Instructional scientists refer to content added to increase emotional interest as seductive details. Note that the seductive details were related to the topic but irrelevant to the instructional goal, which was to build an understanding of weather processes.
College students randomly assigned to the different versions were allowed to study for as long as they wished and then completed an interest rating that asked questions about the emotional and the cognitive interest of the passages. For example, they were asked: "How interesting is this material?" to assess emotional interest and "How much does this material help you to understand the process of lightning?" to assess cognitive interest. Learners were tested with questions that required them to apply what they learned to problems such as: "What could you do to decrease the intensity of a lightning storm?" As shown in Figure 5.7, the base version lacking seductive details resulted in significantly better learning. As you can see in Figure 5.8, learners rated the materials containing seductive details higher in emotional interest and the version that omitted seductive details as higher in cognitive interest. Harp and Mayer (1997) conclude that: "The current findings show that the best way to help students enjoy a passage is to help them understand it" (p. 100).
Mayer, Heiser, and Lonn (2001) replicated these findings in a multimedia lesson in which seductive details similar to those described in the preceding paragraph were added in the form of video clips. Learners who studied lessons with seductive details learned significantly less, with a moderate effect size of .55. Regardless of the delivery media, avoid adding related but extraneous information to spice up technical materials.
Many people prefer to study or work in the presence of background music. Likewise, music is often added to multimedia programs for the same reasons that seductive text and visuals are added—to increase lesson appeal. What is the effect of added audio either in the form of music or sounds related to the content? Here we review evidence regarding the effect of background music on learning, reading, and writing.
WHAT THE RESEARCH SAYS ABOUT QUIET LEARNING AND WORK ENVIRONMENTS
Moreno and Mayer (2000) conducted two experiments that compared learning from multimedia lessons with and without extra audio. The base lessons used narrated animation to explain how lightning forms or how brakes work. The expanded lessons added either instrumental music, relevant environmental sounds, or both music and sounds to the base versions. None of the auditory additions obscured the narration. Following the lesson, learners were tested with questions that required them to apply their understanding of the content. The results from the lightning lesson versions are shown in Figure 5.9. In both lessons, the base version supported learning more effectively than versions with music and sound.
Additional evidence for the negative effects of extraneous audio on learning from reading is reported by Knez and Hugge (2002). They compared learning from a seven-page text read in a quiet environment with reading of the same text in the presence of irrelevant conversational background speech. Recall of text ideas was significantly better among those who read in a silent environment. The authors conclude that irrelevant background noise led to lower reading comprehension as a result of divided attention and/or from competition for limited working memory resources.
Along similar lines, Ransdell and Gilroy (2001) found that the writing of essays in the presence of music (vocal and instrumental) took longer than when writing in a quiet environment. The researchers conclude that music can, like background speech, disrupt writing fluency. To maintain quality of writing, writers slow down their production in the presence of background music. As for practical implications, the researchers suggest: "For all those college students who listen to music while they write on a computer, the advice from this study is clear. One's writing fluency is likely to be disrupted by both vocal and instrument music. A reliable reduction in fluency, about fifty words an hour, is likely to be caused by unattended background music. A student's best bet would be to word process their work in silence, or at least try to reduce background sounds" (p. 147).
Taken together, the research supporting Guidelines 8 and 9 suggest that you:
Write concisely using only the number of words needed to convey the content.
Omit nice-to-know information in technical lessons.
Start technical lessons by building a qualitative mental model before adding quantitative details.
Eliminate visual or audio materials included to add interest.
Study, read, or write in quiet environments.
It is a common misconception that explaining content with multiple modes such as with text and audio narration of that text helps learning. One rationale for this misconception is that by using multiple modes of expression such as on-screen text and audio narration of that text you accommodate both visual and verbal learning styles. Counter to this folk wisdom, we have evidence that multiple content expressions actually overload working memory and depress learning!
Redundancy in training refers to providing more expressions of content than needed for understanding. For example, in our airline safety card shown in Figure 5.10, the illustrations are clear and understandable in themselves. Adding explanatory text would be redundant. Or in a screen from our Excel overloaded example on the CD shown in Figure 5.1, the visual is explained by audio narration to leverage the modality effect. By adding on-screen text that replicates the audio narration, we have created a redundant expression of content. In Guideline 10 we review research related to redundancy as it applies to visuals alone, to narrated visuals and text, and to training of computer applications.
A visual may be self-explanatory for one of several reasons. In some cases, such as the airplane safety card, the drawing itself is inherently self-explanatory to anyone on the aircraft. In other cases a drawing may include some brief embedded text that makes the drawing understandable. Finally, a graphic that is not inherently self-explanatory may be familiar to a specific audience with previous experience using that graphic. In all of these situations, adding words in the form of either text or audio narration is redundant and will depress learning because working memory is better off with just the minimum amount of information needed for understanding. As with all cognitive load management guidelines, redundancy is most detrimental when high demands are made on working memory, such as when the content is complex and the learners are novices.
WHAT THE RESEARCH SAYS ABOUT WORDS THAT ARE REDUNDANT TO ILLUSTRATIONS
Chandler and Sweller (1991) compared learning and the instructional time needed to study three versions of a lesson on blood flow through the heart. The diagram alone version consisted of a cross section of the heart with the chambers labeled and arrows illustrating the direction of blood flow. The other two versions added textual descriptions such as "Blood from the upper and lower parts of the body flows into the right atrium," either embedded in the diagram or as a series of statements placed underneath the diagram. As you can see in Figure 5.11, the diagram alone version resulted in much better learning in one-half to one-third of the time of the other two versions. In this situation, the diagram with embedded arrows and labels was self-explanatory and adding text slowed instructional time and reduced learning.
Leahy, Chandler, and Sweller (2003) contrasted learning from audio explanations of two versions of a temperature graph. One version was self-explanatory and the other was not. For the fifth and sixth graders in the experiment, the basic graph shown in Figure 5.12 is not understandable and requires explanatory words. The second version, shown in Figure 5.13, rendered the graph self-explanatory by embedding short text captions. In the first experiment the graph alone (which was not self-explanatory) was explained by words delivered either by text on the page or by audio narration. Learners were allowed to study the lesson versions for as long as they wished and were then tested with easy and with more complex graph problems. Consistent with the modality effect, the version that explained the graph with audio narration resulted in better learning than textual explanations on complex but not easy test questions.
In the second experiment, learning from the self-explanatory version of the graph (Figure 5.13) alone was compared to learning from this graph with the addition of audio explanations. However, as shown in Figure 5.14, adding audio explanations to the self-explanatory version of the diagram depressed performance on more complex but not easy test questions. For the self-explanatory version of the graph, the words delivered by audio became redundant. The researchers conclude that: "If auditory explanations are used concurrently with, for example, a diagram, which contains sufficient information to be understood alone, the dual-mode duplication of information is redundant and may hinder learning" (p. 414).
From this research, we see that audio explanations aided learning only when the tasks were more complex and only for visuals that were not self-explanatory. Therefore, you cannot universally apply a rule such as "Explain visuals with audio narration." Instead, you must consider the complexity of the task and the meaningfulness of the visual. Of course, the extent to which a visual is self-explanatory can depend on the experience of the learners. We discuss this issue next.
WHAT THE RESEARCH SAYS ABOUT NOVICE AND EXPERIENCED LEARNERS
Kalyuga, Chandler, & Sweller (2000) showed that a visual that was most effectively explained by words presented by audio narration in the beginning stages of learning was most effective when presented without any words during later stages of instruction. A complex diagram like the one shown in Figure 5.15 was presented to learners with explanations delivered with audio narration or without any words over three stages during which learners gained expertise with the diagram. As you can see in Figure 5.16, the version that led to best initial learning lost its effectiveness over time. By Stage 2 there was no difference in learning from the two versions, and by the final stage, the version with the diagram alone was the better choice, both for learning and for instructional efficiency.
Taken together, the research supporting Guideline 10 suggests that:
Visuals may be self-explanatory because they visually incorporate all needed information; text is added to the visual; or the audience is familiar with the visual.
When visuals are self-explanatory, adding more explanations in the form of text or audio narration depresses learning.
When a visual does require an explanation, use audio (modality principle) or integrated text (to avoid split attention).
If you have to design training for a mixed audience that includes both experienced and novice learners, we recommend violating the split-attention guideline and placing explanatory text underneath the diagram. Although it adds a split-attention processing burden to novice learners, this format will allow more experienced learners to easily bypass the text and avoid a redundancy effect. Alternatively, if using audio narration to explain an on-screen visual, provide options for the audio to be suppressed by more experienced learners.
In e-learning, it's common practice to include a combination of on-screen text and audio narration of that text to describe an on-screen visual. For example, in Figure 5.1 from our overloaded Excel lesson demonstration on the CD, a visual of a spreadsheet is explained by on-screen text and audio narration of that same text. This presentation is redundant. A similar redundancy occurs in the classroom or synchronous e-learning when the instructor projects a slide containing a visual and text and reads the text to the class. Incorporating redundant words in visual and auditory form is based on misconceptions that receiving the same message twice will enhance learning and/or that some learners benefit more from visual representations and some from auditory representations so it's best to provide both. Both of these assumptions ignore working memory capacity limits.
Evidence we summarize below consistently suggests that, when describing a graphic, a single representation of words—either in text or in audio—leads to better learning than a dual representation. In our Excel Load Managed asynchronous e-lesson on the CD, on some screens we use just audio narration to explain the content and on other screens we use just text. In contrast, in our overloaded lesson, we violate the redundancy principle by using both text and audio.
WHAT THE RESEARCH SAYS ABOUT OMITTING REDUNDANT TEXT
Figure 5.17 shows a complex technical graph called a Fusion Diagram used by Kalyuga, Chandler, and Sweller (1999) in an experiment that contrasted learning to interpret the diagram from explanations presented in audio, text, and audio plus text. Learners were randomly assigned to study the three different self-paced lesson versions. Learners then rated the difficulty of the lesson and were tested. Reflecting the modality effect, the version that used audio alone to describe the diagram was the most efficient. In contrast, the redundant version that combined text and audio was the least efficient. Figure 5.18 illustrates the efficiency diagram for the three lesson versions. The research team concluded that: "It is widely believed (including among many multimedia instructional designers and educational software manufacturers) that duplicating informationally identical audio and visual material facilitates learning. The results of this experiment suggest that dual-mode presentation under such circumstances has negative rather than positive effects" (p. 352).
Figure 5.18. Diagrams Explained by Audio Are More Efficient Than Diagrams Explained by Text or by Text and Audio.
Mayer, Heiser, and Lonn (2001) reported similar results when comparing learning from the weather lesson that described animated narrations with audio alone, or with audio plus either full or summary text. The versions that added on-screen text (in either full or summary forms) resulted in less learning. The research team calculated an effect size of .53, indicating moderate practical significance.
Some very recent experiments have evaluated the effects of first explaining a visual with audio narration followed by displaying the narrated words in text. By displaying the text after the narration, it is possible that a negative redundancy effect is reduced or even eliminated!
WHAT THE RESEARCH SAYS ABOUT AUDIO FOLLOWED BY TEXT
If on-screen text is displayed on the screen after completion of the audio narration, the negative effects of dual modes are minimized (Kalyuga, Chandler, & Sweller, 2004). A simultaneous audio and text explanation of a complex diagram like the one shown in Figure 5.17 resulted in lower efficiency than a non-concurrent version in which the text was displayed after the audio had played. In both versions each step was presented one at a time and the learner could study the text for as long as he wished before moving to the next step. In a second experiment, the instructional presentation time was preset, eliminating differences in study time. As shown in Figure 5.19, under instructional pacing, the non-concurrent version that presented text after audio resulted in better learning with a large effect size of 1.18 for instructional efficiency.
When explaining a complex visual in asynchronous e-learning, you may want to first present the audio followed by on-screen text. When using this technique, the textual presentation provides a review opportunity for those wanting more study time but can be quickly bypassed by those not needing the additional support.
In the research discussed under Guideline 10, words were used to describe a visual that was not self-explanatory. In other words, the visual required a verbal description. The findings recommend that, when describing such visuals, you should stick with audio alone rather than audio plus identical text presented simultaneously. What do we know about the effects of narration of on-screen text when there is no other visual? For example, in the classroom a PowerPoint® slide presents a sentence of text and the instructor reads that text to the class. Alternatively, an e-learning screen displays three sentences and audio narration reads the same words.
WHAT THE RESEARCH SAYS ABOUT NARRATING TEXT
Two recent studies compared learning from content presented by text alone with content presented by text that was also narrated. As shown in Figure 5.20, the results from these two studies were opposite. Moreno and Mayer (2002) compared learning of college students from two versions of lessons on how lightning works. One version presented the words in audio alone, while the second version narrated the on-screen text. As shown in Figure 5.20 (left side), the version with text and audio resulted in greater learning on a transfer test. In contrast, Kalyuga, Chandler, and Sweller (2004) compared learning of adult trade apprentices from two lesson versions (audio alone and text plus audio) that included a discussion of around 300 to 400 words on four technical topics. As you can see in Figure 5.20 (right side), in this study the versions with audio alone resulted in better learning. What might account for these different results?
A close look shows that in the Moreno and Mayer experiments the words were presented in short segments with pauses between each segment. In contrast, in the Kalyuga, Chandler, and Sweller experiment, the words were presented in one long segment without any pauses. No doubt in the Kalyuga lessons, a lengthy text presented without pauses imposed greater load than did the shorter texts used in the Moreno and Mayer experiments. A long audio segment is difficult to hold and process in working memory and even more difficult to coordinate with visuals, resulting in a heavy load.
We recommend that you can use audio and identical text when presenting short content segments. However, a negative effect may occur if you narrate lengthy text segments without pauses. We need additional research to guide decisions regarding narration of text in the absence of visuals.
In Chapter 4, we recommended that you integrate step-by-step computer instructions onto screen simulations of the application and present the instruction wholly on the computer. Often new computer application training programs will present the steps and visuals on the computer and replicate them in a manual for reference. Having exactly the same content duplicated in two media, that is, on the computer and in a manual, is a redundant expression of content and will lead to less efficient learning!
WHAT THE RESEARCH SAYS ABOUT REDUNDANT DELIVERY MEDIA
Learning is better when words are placed on a simulation of spreadsheet screens than when words are presented in a manual (Cerpa, Chandler, & Sweller, 1996). The lesson contained partially in the manual and partially on the computer caused learners to split their attention between the manual and the computer screen. A third lesson version in the Cerpa, Chandler, and Sweller experiment displayed the words on a simulation of the application and, in addition, gave the learners a copy of all of the training screens in a manual. Thus the learners had two duplicate versions—one on the computer and a second in a manual.
As you can see in Figure 5.21, for more complex tasks only, the version that incorporated everything on the computer resulted in better learning than the version that added a manual that duplicated the training screens. The self-contained computer version was also better than the split attention version in which the text was placed in the manual and the learner used the computer to apply the steps. In short, formats that led to either split attention or redundancy diminished learning in materials designed to teach computer applications. Similar results using materials to train different applications were reported by Sweller and Chandler (1994) and Chandler and Sweller (1996).
An optimal format for learning computer applications is one that incorporates all instruction into the computer screens and omits either a redundant version that duplicates the screens in a manual or a split attention version that includes textual directions in the manual. In short, drop the manual.
There are solid research and psychological reasons for recommending that you offer less rather than more in your training by:
Writing concise text to present your content
Omitting words, visuals, or audio that do not contribute to understanding
Building a qualitative mental model first in technical lessons and adding quantitative details later
Avoiding redundant words, modalities, and delivery media
The benefits of a less-is-more approach can be realized in decreased time for development of instruction, reduced learning time, and better outcomes. In short, a lean approach is an efficient approach!
Chapter 5: Weed Your Training to Manage Limited Working Memory Capacity. John begins by defining the redundancy effect followed by a discussion of repetition versus redundancy, redundancy in computer training, and redundancy in instructor presentations.
Compare our Before: Overloaded Excel Web-Based Lesson to the After:
Load-Managed Excel Web-Based Lesson for the following elements:
The use of audio and text: Redundant in the Before version and not redundant in the After version.
Insertion of extraneous animation and content in the Before version removed from the After version.
All content is included in the computer lesson; no reference to an adjunct manual is made.
Notice that in the synchronous e-learning virtual classroom example:
Instructor does not read on-screen text to the class.
All content is included in the computer lesson.
The lesson content is succinct and relevant to the instructional goal.
In the past three chapters we have focused on ways to manage cognitive load by (1) making best use of the visual and phonological centers in working memory via the modality principle; (2) maximizing limited working memory capacity by focusing attention and avoiding split attention; and (3) streamlining lessons in ways that take a minimalist approach to instructional materials and interfaces.
In the next chapter we suggest that in some situations you can manage load by providing well-designed working aids to support learning and performance.