Part 1:
The Instructional Role of Illustrations

Illustration Functions

One of the more useful ways of approaching instructional illustrations is by examining their functions (Duchastel, 1978). In this discussion, we will examine how illustrations can attract attention, aid retention, enhance understanding, or create context. For example, showing a photo of a dramatically beautiful cloud image at the opening of a presentation on climatology does more to attract attention than explain the content. Providing a table of cloud type images organized according to the cloud classification scheme might be seen as primarily aiding retention. Drawing a diagram illustrating the processes in the formation of a particular cloud type will help explain these processes for increased understanding. If we want to place weather forecasters in an instructional scenario where they are asked to practice making forecast decisions, we might create context by providing a user interface that resembles a forecast office with a window revealing storm clouds growing outside.

Illustrations don't always fall neatly into one of these categories. Often they will serve two or more of these functions. Opening with a dramatic cloud image may grab attention and at the same time help explain or make memorable the cloud structure you are discussing. For this section, we will use the categories of attention, retention, understanding, and context creation to organize a general discussion of the benefits to using illustrations in instruction.

titlegfx.jpg (38371 bytes)

This module title graphic serves mainly to be attractive and gain attention (in addition to identifying  the instructional module, of course). This is an important role, however. One could argue that it also begins to establish a context by introducing abstract images of the ice crystal and pilot report codes.

pyramid.gif (5316 bytes)

Using the simple shape of a cone, this illustration memorably represents the relative times spent by a weather forecaster with the various scales of data. Such a device can aid tremendously in a student's retention.

cldform.gif (24267 bytes)

This illustration helps learners to better understand various processes of cloud formation.
When we are considering illustrations as aids to retention and understanding, we also need to look at how illustrations can provide performance support. While the use of illustrations as performance support is not the focus of this discussion, it is appropriate to touch on those aspects in this section as well. After all, task practice (with performance aids) is obviously a valid part of instruction. And if we blur the definition of instruction a bit to include the long path toward gaining expertise that follows formal instruction, then performance aids can also be seen as instructional.


Gaining the attention of students is a precondition to any kind of learning. It appears first in most lists of events in the instructional process (e.g., Gagné & Briggs, 1979) both because it is a precondition and because it can be too easily slighted among the more "constructive" aspects of instruction. In the terms of cognitive psychology, gaining attention is critical because of the limited capacity and duration of the "short-term" or "working" memory (which has a capacity of five to nine items and a duration of 10 seconds, according to most researchers). Working memory is the place where conscious mental work is performed. If new items need to be brought into working memory, others must be dropped. If items are not used or rehearsed, they tend to be forgotten. If they have been used sufficiently, they will be stored in long-term memory for later recall.

According to cognitive theory, only a small fraction of all sensory stimuli that reach the brain actually make it to the working memory, a process known as selective perception (Gagné, 1985).  Considering the limitations of working memory, it is easy to understand why such a process is necessary. Without it we would be inundated with stimuli in an instant. With all the stimuli that are vying for attention, it is a challenge to keep learners focused on the intended message or task. Using illustrations can help because they can be complex to the senses, and because they can provide novelty in the probably more plentiful stream of spoken words or written text.


Complexity draws and holds attention (Fleming & Levie, 1978). Illustrations, because they can display varieties and interrelationships of lines, shapes, colors, spaces, and text, tend to be more complex to the senses than lectures or text alone, and so they will stand out to the learner. It should be added, however, that there are limits to the benefits of complexity. If complexity exceeds the processing capabilities of learners, they may withdraw attention in self-defense. There is an optimal level of complexity that varies with the age and ability of learners that must be taken into consideration.

complex.jpg (4218 bytes)

Our attention is drawn to the circle on the right because of the higher complexity introduced by multiple colors.

Novelty is also one of the keys to gaining attention (Fleming & Levie, 1978). Our attention is drawn to what is new or unexpected in our sensory field. This is the reason police and other emergency sirens use rapidly changing pitches rather than sheer volume to announce the need for caution. Even though the ideas you present may be quite interesting or complex, a long, uninterrupted lecture or pages of text may introduce sensory monotony. Judicious use of illustrations within such presentations will introduce novelty and help hold the attention of learners. As with complexity, however, instructors should not over emphasize novelty as a strategy. Anything that is originally novel and attention-getting can become monotonous with overuse.

textonly.gif (61839 bytes)

textgraf.gif (61476 bytes)

One of the benefits of using illustrations is simply the visual variety they produce. The lower screen is much more inviting than the one above and is more likely to engage students.

complex2.jpg (7869 bytes)

The attention gained by the more complex circles is now lost due to overuse. The more simple, solid color circle now stands out because of its novelty.

Retention usually refers to keeping information available in the long-term memory, but the ability to retain needed information in working memory is equally important to learning and performance. Problem solving, for instance, requires keeping many factors of a situation in mind while analyzing them for a solution. Illustrations provide assistance for both kinds of retention.

Long-Term Retention

One of the basic ways that illustrations aid retention relates to the well-researched (but not undebated) dual-coding theory of memory (Paivio, 1971). This theory proposes that information is stored in long-term memory both as verbal propositions and as mental images. It suggests that when information is presented verbally and visually it has a better chance of being remembered. Corroborating research shows that concrete words are remembered better than abstract words, and that pictures alone are remembered better than words alone (Fleming & Levie, 1978). From the dual-coding perspective, an explanation is that concrete words help us generate associated mental images, and that pictures alone help us to generate associated words, in addition to detailed mental images. The combination of verbal proposition and mental image establishes multiple pathways by which the information can be retrieved from memory.

frog.gif (22897 bytes)

frogconc.jpg (46141 bytes)

The fact that we can store the concept "frog" as both a verbal proposition and a mental image increases the likelihood that we will recognize one when we see or read about it.
Simultaneous verbal and visual representation probably does not completely account for the memorable nature of images, though. It may be that, because they are concrete, images are intrinsically more durable in memory than verbal propositions. The extraordinary aberration of photographic (or idetic) memory, in which people retain incredibly detailed images in memory after a single viewing, has no corresponding aberration in verbal memory. (Extraordinary "aural" memory has been noted, however, in people such as Mozart.) Feats of seemingly remarkable verbal memory, such as Greek and Roman orators and storytellers reciting lengthy speeches or enormous epic poems from memory, actually use visualization strategies (linking words to mental images) to enhance recall.

Obviously, illustrations can help learners retain information by stimulating the mental images to facilitate dual-coding. They can be of special value when teaching abstract information by providing images that learners might not generate on their own. Charts showing trends, diagrams of processes, and visual metaphors of abstract ideas are all ways of providing images for abstract ideas.

cncpmap3.gif (56464 bytes)

The complex relationships of the atmospheric properties and processes that determine thunderstorm types are memorably illustrated in this "concept map." These relationships are difficult to sum up as succinctly in words.

primates.gif (12810 bytes)

Illustrations can help generate associated mental images for abstract information. Learners may remember the primate taxonomy better if they can call up this spatial arrangement from memory.
Another way of looking at how illustrations help students retain abstract information is their ability to assist in building well-structured "schemas." Cognitive theory proposes that we store knowledge in networks of interrelated propositions called schemas (Norman, 1982). Recall is achieved by following a path of related propositions until the desired information is encountered. Complex and accurately structured schemas facilitate recall by providing many alternate and efficient paths to reach information.

filecab.gif (15612 bytes)schema1.gif (8456 bytes)

Psychologists prefer to think of our knowledge as stored in multidimensional proposition networks, or schemas, rather than using the more linear "file cabinet" metaphor.
Graphic organizers are diagrams that explicitly reveal a useful schema (Jonassen & Hawk, 1984). Presenting these organizers to students, either in advance of instruction or as summaries, facilitates the process of "organization," the structuring of a schema. Several studies also show the advantages of having students finish incomplete graphic organizers, or requiring students to create their own (e.g., Geva, 1983). Because participatory graphic organizers require the additional involvement of students, this strategy may have an even more positive effect on recall than instructor-generated organizers (as long as feedback is provided to correct possible misconceptions). Other related graphical tools for building schema include pattern notes (usually student-generated), concept maps, and all varieties of diagrams.

rocks1.gif (19087 bytes)

This is an example of a graphic organizer for a unit on rock classification in a geology class. Using graphic organizers may help structure useful schemas in the minds of students. They can provide the scaffolding from which to build more detailed knowledge.

rocks2.gif (15763 bytes)

"Participatory" graphic organizers, which students must complete themselves, may work best of all in helping students build resilient memory structures.
Retention in Working Memory

Illustrations can also be seen as assisting the short-term or working memory by making more information readily available. Illustrations can present simultaneously all the information needed to explain a topic or perform a task. Where a linear string of words must use a series of semantic cues to its organization over the course of its passage, an illustration can use lines, boxes, arrows, space, color, typefaces, and the relative distance between elements to communicate information about the relationships of those elements. Because the reader can see this information at a glance or with minimal study, graphical presentation can be more efficient than words alone (Winn, 1987). For example, charts with multiple columns and rows can reveal the complex relationships between large amounts of information. Such information would be difficult to present and even more difficult to comprehend in words alone. When students read prose or hear exposition, they have to hold information in working memory long enough to relate it to information presented later– a difficult task in a long passage. Simultaneous presentation can reduce the processing load on the working memory and thus help students better see relationships within the information.

Illustrations also assist working memory by helping to create hierarchies of information. Since the capacity of working memory is so limited, the mind uses hierarchies to reduce the numbers of necessary details it has to keep in store. A visual representation or a label of a subset of information can act as a pointer. This pointer can be stored in working memory instead of all its elements to reduce the mental workload, yet the pointer can keep the elements readily available for quick reference if needed (Gagné, 1985). Because illustrations can present a large amount of information simultaneously, and can use various visual cues to suggest hierarchical groups, they can become powerful pointers. The Greek and Roman speakers mentioned above sometimes used a strategy that involved associating passages of the poem or speech with images of objects and rooms in a real house. When reciting the poem, they would imagine themselves exploring that house, and would recall passages into working memory as they "visited" each room filled with reminders (Cicero, On the Good Life, as cited in Hilts, 1995). This type of strategy is so effective that even when the elements within a subgroup are only tenuously, or even artificially related, the visual pointer can solidly link them.

hierarchy.jpg (56805 bytes)


Heirarchies help learners retain information by "chunking" it. A heirarchy illustration provides a big picture, but one that is conveniently broken into categories. Recall the category and you may recall the details more easily.
Distributed Cognition

Illustrations sometimes aid retention, or even circumvent the need for retention, by substituting for it. "Knowledge in the world" (Norman, 1988) is information that we can refer to in outside sources (reference materials, instructions, control labels) rather than commit to memory ("knowledge in the head"). When practical, it is more efficient to leave knowledge in the world, temporarily or permanently, rather than to exert effort to memorize it.

The concept of "distributed cognition" (Norman, 1993) suggests that we not think of ourselves as isolated minds. Rather, we should consider how teams of people function together to perform tasks, and how our tools and work environment act as extensions of our cognition and contribute to our ability to perform. In other words, we should consider how knowledge in the world and in the head work together and work to improve the relationship. Illustrations play a key role in our distributed cognition.

For example, "job aids" are sometimes illustrations of procedural steps or physical layouts that are designed to be read while performing a procedure. Common examples are the step-by-step assembly instructions included with products sold unassembled. In this case, the procedure is intended to be performed only once and never memorized, but job aids are also useful for learning procedures that eventually must be performed automatically. Decision trees (see example immediately below) and flow charts of decision-making procedures help us apply a decision-making strategy by reminding us what questions to ask and what data to pay attention to. After time, we may internalize the process and no longer need to refer to the decision tree.

hydflow.gif (19345 bytes)

Students must first learn the steps of a complex procedure as verbal knowledge (facts and rules that are recalled explicitly, achievable via rote learning) before they become capable of performing it automatically. Yet repeated practice is the most effective way to learn to perform the steps and their subtleties (learning by doing). Referring to job aids can help students through the initial stage by letting them practice complex procedures without committing large amounts of data to memory (which at first may seem meaningless). When they have performed the procedure many times, the data begins to acquire meaning, making it more easy to remember. In other words, the memorization can take place when it is most effective.

We’ve discussed how illustrations gain attention and make information more memorable, but how do they make information easier to understand? One answer seems to lie in the ability of illustrations to make the abstract more concrete, and to make the concrete more abstract. Another answer has to do with calling upon alternative, under-utilized mental skills.

Illustrations for Concrete Information

Illustrations are obviously valuable in teaching concrete concepts and how to work with concrete objects. It would be nearly impossible to teach students how to classify cloud types without showing illustrations of each type and pointing out their differences. It would be equally difficult to teach medical students human anatomy without illustrations. It may be stating the obvious, but illustrations are also critical for showing those things that can’t normally be seen, such as the interiors of things and the microscopic and macroscopic. Illustrations are important for teaching principles that have spatial determinants. An illustration would be especially appropriate for showing that different species of trees dominate the ecosystem at different altitudes.

sm2cm30.gif (14418 bytes)

Illustrations are especially important for teaching spatial concepts and principles.

structure1.gif (52368 bytes)

Illustrations can make the invisible visible. Cross sections are important for showing internal structure (in this case, a hurricane). Arrows can serve a lot of purposes, such as revealing the paths of the wind.

therm.gif (8168 bytes)

Colors are a useful tool for showing temperature differences.
Illustrations for teaching concrete information are, in most cases, more effective if they are somewhat abstracted. Overly realistic illustrations, including photographs, may present too many unnecessary details that can obscure the critical elements. Line drawings or other simplifications, which suggest the real object but highlight the critical elements in an unrealistic way, are usually more effective than realistic illustrations. Of course, realism may be necessary for the later stages of developing expertise, where objectives require students to identify or manipulate objects in the real world. Proponents of situated learning approaches also warn us not to abstract learning excessively from real world experiences.

osf21300.gif (15535 bytes)

ilxpirep.gif (21655 bytes)


Weather forecasters must work with data and imagery that can be cryptic to them (let alone the rest of us) because it is so abstracted from reality. Instructional illustrations can provide a somewhat more realistic depiction of what they see in that data, helping them visualize its physical implications. While they can't rely on such images on the job, use of more realistic illustrations during training can be a stepping stone to better use of the data.
Illustrations for Abstract Information

Illustrations prove valuable in teaching abstract information by providing spatial metaphors for logical structure. Using space, lines, boxes, arrows, color, and the relative distance between elements, an illustration can provide a concrete equivalent of abstract ideas. Many writers have outlined methods of making graphic representations of the common structures of information, such as comparison and contrast, procedural steps, description, causation, and chronology (e.g., Jones, Pierce, & Hunter, 1989). For example, column charts are efficient for comparing or contrasting the characteristics of items; time lines are good at showing chronological relationships; flow charts can show causation clearly.

continum.jpg (7162 bytes)
compare.jpg (12574 bytes)
cycle.jpg (12055 bytes)

Here are some examples of common templates that can be used for structuring information graphically.

1mfrclod.gif (24970 bytes)

This is a nice example of an illustration that uses the continuum template.

correla1.gif (13910 bytes)
correla2.gif (13737 bytes)
correla3.gif (13693 bytes)

This series of very simple graphics provides a tangible expression of the statistical concept of "correlation."

ensmfcst.gif (21535 bytes)

There are advantages in averaging an ensemble of computer model runs rather than relying on a single run based on a best guess of the initial state. This abstraction makes the point clear by reducing the problem to lines and points.
Data graphics are a special case of illustration that provide spatial representations of numeric data. Data graphics, such as bar graphs and pie charts, are indispensable for revealing trends or other relationships of data that would not be easily discerned if the data were presented only numerically.

ws_klmp5.gif (9107 bytes)

Data graphics help reveal trends, comparative quantities, or areas of overlap indicated, but not made obvious, in the raw numerical data.

ch367.gif (27609 bytes)

It can be beneficial at times to combine data graphics and conceptual graphics. The conceptual graphic can provide a physical representation of the principle revealed in the data. Here we see that at particular wavelengths, surface temperatures, as indicated by energy emitted by water molecules near the earth's surface, are not seen by weather satellites.
The omnipresence of data graphics in research papers, newspapers, corporate reports, etc., shows that illustrations are important not only in learning abstract information, but also in the ongoing effort to manipulate it. We all use illustrations as performance aids. It has been said that "solving a problem simply means representing it so as to make the solution transparent" (Simon, 1981). We draw organizational charts, process maps, Gantt and PERT charts, etc. to help us manage information and see what courses of action are necessary.

offlow1a.gif (4076 bytes)

Using graphic vectors we can explain how you can estimate the orientation and speed of ground-relative wind gusts from a storm outflow as they interact with the environmental winds--a confusing problem otherwise.
Alternative Mental Skills

Another advantage to using illustrations is that they address alternative mental skills or learning styles of students. In our culture we tend to stress verbal over visual learning. Because the written text and the lecture are the primary modes of instruction, students have fewer chances to use and develop visual learning abilities. Tapping these abilities with illustrations may reinforce learning in ways that verbal learning alone would not.

This idea recalls "whole-brain learning," which some researchers have advocated (Toth, 1980).  Left-brain/right-brain theory proposes that the left hemisphere of the brain is best at verbal and other symbolic processes, and that the right hemisphere is better at visual and creative processes. By tapping both hemispheres in whole-brain learning, each hemisphere can contribute its own strengths, and more thorough learning may take place. The idea that it is beneficial to address multiple learning styles (visual vs. aural, etc.) is based on a similar concept, although here it is proposed that each learner has a predisposition to a particular style. If we use only one teaching mode, such as lecture, we may be leaving behind those students with equal ability, but an alternate preferred, and presumably optimal, learning style.

Creating Context

Situated Learning

When theorists speak of "situated cognition," they are referring to the way our ability to perform is closely tied to how well we become enculturated in the social and physical environment of practice. They suggest that "communities of practice" create their own situation-specific meanings, and that cognition in any specific "community of practice" is intertwined with one's interactions with colleagues and the tools-of-the-trade (Lemke, 1997). This point-of-view suggests that knowledge cannot be effectively objectified and taught isolated from the environment in which we want learners to perform.  Instead, we need to offer opportunities for practice in those environments, or at best, in close approximations of them.

Creating "case-based learning" (Schank & Cleary, 1995) materials, through computer-based instruction or some other medium, can fulfill the need for situated learning. Someone who has been exposed to a rich variety of cases (such as an experienced expert) will be able to draw from them when they face analogous situations. Examples of case-based or scenario-based learning tools can range from collections of stories culled from experts to highly interactive instructional simulations. Illustrations within such materials can play an important role in establishing a realistic context. They can provide representations of the environment and tools of practice, adding realism to the cases or simulations that can act as stepping-stones to actual performance. For example, providing a computer interface whose look simulates the actual data-display system of a weather forecaster may facilitate the transfer of computer-based forecaster training to the job by creating the proper mindset for its application.

dmpliss2.gif (71103 bytes)

dmpliss4.gif (71220 bytes)

dmpliss7.gif (70528 bytes)

Lessons that use illustrations to suggest the work environment may aid in "situating" the learning appropriately. These graphics were used in various stages of a simulation created for emergency managers. They help to create an increasingly stressful atmosphere and make the simulation more realistic (in addition to adding a touch of humor).

icse2812.gif (20719 bytes)

If we want meteorologists to be able to better transfer training to their jobs, it is important that at least a portion of the instructional graphics represent the type of data graphics that are used on the job. We should resist always trying to "improve" the data by making it more readable, or we risk not improving performance.
Instructional Metaphors

Useful instructional contexts can be metaphorical or analogical as well as realistic. These types of contexts may also help make information better understood by linking new knowledge to existing knowledge that has an analogous relationship, or simply by  providing a mnemonic device. Illustrations can make effective mnemonic devices.

skater.gif (20203 bytes)

The stretching of vorticity is a difficult physical concept, but comparing it to a skater's conservation of momentum helps to clarify what we mean.

puz4.gif (41103 bytes)

This content menu is constructed in the form of a jigsaw puzzle. The content of this instructional module (Numerical Weather Prediction) has no relationship to a puzzle except that it is often puzzling to its users. The puzzle metaphor may be attention-getting and memorable nonetheless.

Duchastel, P.C. (1978) Illustrating Instructional Texts, Educational Technology, Nov. 1978.

Fleming, M. & Levie, W. H. (1978) Instructional Message Design, Englewood Cliffs, N.J.: Educational Technology Publication.

Gagné, R.M. & Briggs, L.J. (1979) Principles of Instructional Design (Second Edition), New York: Holt, Rinehart, and Winston.

Gagné, E. (1985) The Cognitive Psychology of School Learning, Boston: Little, Brown, and Company.

Geva, E. (1983) Facilitating reading comprehension through flowcharting. Reading Research Quarterly, 18, 384-405.

Hilts, P.J. (1995) Memory's Ghost: The Strange Tale of Mr. M. and the Nature of Memory, New York: Simon & Schuster.

Jonassen, D. & Hawk, P. (1984) Using graphic organizers in instruction. Information Design Journal, 4, 58-68.

Jones, B.F., Pierce, J., & Hunter, B. (1989) Teaching students to construct graphic representations. Educational Leadership, Dec. 88/Jan. 89, 20-25.

Lemke, J.L. (1997) Cognition, Context, and Learning: A Social Semiotic Perspective. In D. Kirshner & J. A. Whitson (Eds.) Situated Cognition: Social, Semiotic, and Psychological Perspectives, Mahwah, NJ: Lawrence Erlbaum Associates.

Norman, D. A. (1982) Learning and Memory, San Francisco: Freeman.

Norman, D.A. (1988) The Psychology of Everyday Things, New York: Basic Books, Inc., p. 55-80.

Norman, D. A. (1993) Things That Make Us Smart, Reading, MA: Addison-Wesley.

Paivio, A. (1971) Imagery and Verbal Processes, New York: Holt, Rinehart, and Winston.

Schank, R.C., & Cleary C. (1995) Engines for Education. Hillsdale, NJ: Lawrence Erlbaum Associates.

Simon, H.A. (1981) The Sciences of the Artificial (2nd Edition), Cambridge, MA: MIT Press.

Toth, M.A. (1980) Figures of thought. The use of diagrams in teaching sociology. Teaching Sociology, 7, 409-424.

Winn, W. and Holliday, W. (1982) Design Principles for Diagrams and Charts. The Technology of Text, (Jonassen, D.H., Ed.), Volume 1, 277-299.

Winn, W. (1987) Charts, Graphs, and Diagrams in Educational Materials, The Psychology of Illustration (Volume 1), New York: Springer-Verlag, 152-198.