when is an illustration worth ten thousand...

Journal of Educational Psychology1990, Vol. 82, No. 4,715-726

Copyright 1990 by the American Psychological Association, Inc.0022-O663/90/SOO.75

When Is an Illustration Worth Ten Thousand Words?Richard E. Mayer

University of California, Santa BarbaraJoan K. Gallini

Department of Educational Psychology,University of South Carolina

In three experiments, students read expository passages concerning how scientific devices work,which contained either no illustrations (control), static illustrations of the device with labels foreach part (parts), static illustrations of the device with labels for each major action (steps), ordynamic illustrations showing the "off" and "on" states of the device along with labels for eachpart and each major action (parts-and-steps). Results indicated that the parts-and-steps (but notthe other) illustrations consistently improved performance on recall of conceptual (but notnonconceptual) information and creative problem solving (but not verbatim retention), and theseresults were obtained mainly for the low prior-knowledge (rather than the high prior-knowledge)students. The cognitive conditions for effective illustrations in scientific text include appropriatetext, tests, illustrations, and learners.

The two major media for communicating scientific infor-mation to students are words and pictures. In spite of thetraditional bias toward verbal over visual forms of instruction,a growing research base suggests that text illustrations canhave important effects on student learning (Levie & Lentz,1982; Mandl & Levin, 1989; Mayer, 1989a; WiUows &Houghton, 1987). This article is based on the premise thattechniques for enhancing students' visual learning of scientificinformation represent a relatively untapped potential for im-proving instruction.

When is an illustration likely to improve student learning?What makes a good illustration? Which features of illustra-tions make them effective? How can we use illustrations toimprove students' learning from expository prose? These arethe kinds of questions that motivate our study. In this intro-duction, we examine the idea that one function of illustrations(which we call explanative illustrations) is to help readersbuild runnable mental models and that several conditionsmust be met for explanative illustrations to be effective.

Cognitive Functions of Illustrations

What makes a good illustration? In order to define whatwe mean by "good illustration" we must further specify "goodfor what?" Levin (1981; Levin, Anglin, & Carney, 1987) hasproposed five functions of text illustrations: (1) decoration—illustrations can help the reader enjoy the textbook by makingit more attractive (but without being relevant to the text); (2)representation—illustrations can help the reader visualize aparticular event, person, place, or thing (such as found com-monly in narrative passages); (3) transformation—illustra-

The authors gratefully acknowledge the assistance of Jeffrey LeeSteinberg in running subjects and Bill Lee in conducting statisticalanalyses. The first author was mainly responsible for Experiments 1and 2. The second author was mainly responsible for Experiment 3.

Correspondence concerning this article should be addressed toRichard E. Mayer, Department of Psychology, University of Califor-nia, Santa Barbara, California 93106, or to Joan Gallini, Departmentof Educational Psychology, University of South Carolina, Columbia,South Carolina 29208.

tions can help the reader remember key information in a text;(4) organization—illustrations can help the reader organizeinformation into a coherent structure; and (5) interpreta-tion—illustrations can help the reader understand the text. Inthis article we focus exclusively on illustrations aimed atserving the interpretation function, which we call explanativeillustrations.

For this study we set our instructional goal as the promotionof learner understanding of how scientific systems work, asmeasured by qualitative reasoning about scientific systems(Bobrow, 1985). We define a "good illustration** as one thatpromotes the reader's understanding of how a scientific sys-tem works. Therefore, we focus our study on explanativeillustrations, that is, illustrations that promote interpretationprocesses.

Based on theories of mental models (de Kleer & Brown,1985; Gentner & Stevens, 1983; Kieras & Bovair, 1984;Larkin & Simon, 1987; White & Frederiksen, 1987), weidentify two major features of illustrations that could helplearners build runnable mental models: system topology andcomponent behavior. System topology refers to the portrayalof each major component within the structure of the system.For example, the parts illustrations shown in Figures 1 and 2present the system topologies of a brake system and a pumpsystem, respectively. In Figure I, the labeled componentswithin the braking system include the tube, wheel cylinder,small pistons, brake shoe, and brake drum.

Component behavior refers to the portrayal of each majorstate that each component can be in and the relation betweena state change in one component and state changes in othercomponents. For example, the parts-and-steps illustrations inFigures 1 and 2 show alternative states of the components ina braking system and a pump system, respectively. In Figure1, the parts-and-steps illustrations show the "before" and"after" states of the tube, smaller pistons, brake shoes, andwheel: the fluid in the tube can be compressed or not, smallpistons can be back or forward, the brake shoe can be pressingagainst or away from the drum, and the wheel can be spinningor still. In addition, the relations among state changes areemphasized in the steps illustrations and the parts-and-stepsillustrations: if the fluid in the tube is compressed, the smaller

715

716 RICHARD E. MAYER AND JOAN K. GALL1NI

PARTS ILLUSTRATION STEPS ILLUSTRATION

TUBEWHEEL CYLINDER

SMALLER PISTON

BRAKE DRUM

BRAKE SHOE

When the driver steps on the car's brake pedal...A piston moves forward inside the master cylinder (not shown)

The piston forces brake fluid outof the master cylinder and throughthe tubes to the wheel cylinder.-

In the wheel cylinder, the increasein fluid pressure makes a set ofsmaller pistons move.

When the brake shoes press againstthe drum, both the drum and thewheel stop or slow down.

PARTS AND STEPS ILLUSTRATION

(TEXT INCLUDES BOTH OF THE ABOVE ILLUSTRATIONS)

Figure 1. Illustrations of a brake system. (Adapted from The World Book Encyclopedia. Vol. 2, p.463, 1987, Chicago: World Book, Inc. Copyright 1990 by World Book, Inc. by permission of thepublisher.)

pistons move forward; if the smaller pistons move forward,the brake shoe will push against the drum; if the shoe pressesagainst the drum, the spinning wheel will stop.

Current theories of mental models suggest the potentialeffectiveness of qualitative models in teaching students aboutscientific systems (de Kleer & Brown, 1985; Kieras & Bovair,1984; Gentner & Gentner, 1983; White & Frederiksen, 1987).Such models are constructed to impart visual representationsof how a system looks and how it functions under variousstate changes. A straightforward implication of this work isthat experience with qualitative models will allow students toacquire alternative conceptualizations of how a system worksand to develop skill in predicting the causal behavior of thesystem. Unfortunately, empirical research in this area is lim-ited, but even more importantly, it generally ignores theeducationally relevant issues of how illustrations can be de-signed and used to promote the acquisition of runnablemental models. Our line of research builds on existing theoriesof mental models, but focuses on the instructional question"When is an illustration most likely to be effective in pro-moting scientific understanding?"

Conditions for Effective Illustrations

As summarized in Figure 3, Mayer (1989b) has proposedfour conditions that must be met for illustrations to be effec-tive in promoting understanding of scientific text: (1) exploit-ative text—the text must present a cause-and-effect systemthat allows for qualitative reasoning (Bobrow, 1985); (2) sen-sitive tests—the performance measures must evaluate thelearners understanding and qualitative reasoning about thesystem; (3) explanative illustrations—the illustrations musthelp the learner build a runnable mental model of the system;and (4) inexperienced learners—the students must not spon-

taneously engage in active learning processes such as theconstruction of a runnable mental model of the system. Insummary, the text, tests, illustrations, and learners must allbe appropriate for the instructional goal of fostering meaning-ful learning.

In this article we report on three experiments that each testsix predictions generated from the model in Figure 3. In eachexperiment, students read a scientific text that contains ex-planative illustrations, nonexplanative illustrations, or no il-lustrations. Then students write down what they rememberand answer questions about the text.

First, we examine two predictions' concerning the testcondition shown in Figure 3.

1. Explanative illustrations improve recall of explanativeinformation but not nonexplanative information. This predic-tion follows from schema theory (Rumelhart & Ortony, 1977):conceptual understanding depends on formulating a schemaor knowledge structure about how a system operates. Specif-ically, the schema describes relations among parts of thesystem including how a change in one part of the systemaffects a change in another part of the system (i.e., explanativeinformation). Supplemental factual information (i.e., nonex-planative information) might temporarily be stored in mem-ory as discrete pieces of knowledge but generally would notbe assimilated within the schema of the system.

2. Explanative illustrations improve creative problem solv-ing but not verbatim retention. This prediction is based on theidea that the most direct measure of conceptual understandingis a problem solving test that requires what Bobrow (1985)calls "qualitative reasoning about physical systems" (p. 1) or

' Predictions 1 through 4 are based on the idea that condition 4 ismet, namely that the learners are inexperienced; therefore, dataanalysis concerning these predictions in each experiment was basedonly on data from low prior-knowledge learners.

WHAT MAKES A GOOD ILLUSTRATION?

PARTS ILLUSTRATION STEPS ILLUSTRATION

717

HANDLE

• PISTON

INLET VALVE

OUTLET VALVEH06E

As the rod is pulled out,air passes through the pistonand fills the areas betweenthe piston and outlet valve.

As the rod is pushed in,the inlet valve closesand the piston forces airthrough the outlet valve.

PARTS AND STEPS ILLUSTRATION

t\

•HANDLE

As the rod is pulled out,

-air passes through the piston

PISTON

INLET VALVE

OUTLET VALVE

and fills the area between thepiston and the outlet valve.

- As the rod is pushed in

• the inlet valve closes

and the piston forces air throughthe outlet valve.

Figure 2. Illustrations of a pump system, (Adapted from The World Book Encyclopedia, Vol. 15, p.794, 1987, Chicago: World Book, Inc. Copyright 1990 by World Book, Inc. by permission of thepublisher.)

what de Kleer & Brown (1983) call "running a causal model"(p. 158). In contrast, verbatim retention is irrelevant to thegoal of meaningful understanding.

Second, we examine two predictions concerning the illus-tration condition shown in Figure 3.

3. Explanative illustrations are more effective in improvingconceptual recall than nonexplanative illustrations. This pre-diction aims to isolate the features of effective illustrations,namely, only illustrations that explicitly concretize both the

system topology and component behavior are likely to fosterthe learner's building of a runnable mental model. Learnerswho have built a runnable mental model are more likely torecall information about how one state change affects anotherbut less likely to remember irrelevant facts, as compared tostudents who have not built mental models.

4. Explanative illustrations are more effective in improvingproblem-solving performance than nonexplanative illustra-tions. Following the same line of reasoning as presented for

718 RICHARD E. MAYER AND JOAN K. GALLINI

Explanative x - (nonexplanative text)text? /

Experiment I (Brakes)

Method

(nonconceptual test>

(nonexplanative illustrations)

(high prior-knowledge learners)

Figure 3. Four conditions for effective illustrations.

Prediction 3, learners who possess runnable mental modelsshould be able to engage in qualitative reasoning better thanlearners who do not (White & Frederiksen, 1987).

The final two predictions are based on the learner conditionin Figure 3.

5. Explanative illustrations will improve recall of explana-tive information for low prior-knowledge learners but not forhigh prior-knowledge learners. This prediction is based on theidea that high prior-knowledge learners come to the learningsituation with a repertoire of techniques for strategically usingtheir domain knowledge of mechanical systems (Alexander &Judy, 1988; McDaniel & Pressley, 1987), whereas low prior-knowledge learners do not. High prior-knowledge learnerspossess strategies for visually representing information in thetext and can coordinate imagery strategies with other proce-dures. These techniques help high prior-knowledge learnersfocus on the explanative information during learning so spe-cial illustrations are not needed. In contrast, imagery strategyresearch (McDaniel & Pressley, 1987) demonstrates that lowprior-knowledge learners, who lack skill in visualizing on theirown, will be more likely to profit from imagery-based instruc-tional supplements than high prior-knowledge learners.

6. Explanative illustrations will improve problem-solvingperformance for low prior-knowledge learners but not for highprior-knowledge learners. Following Prediction 5, high prior-knowledge learners spontaneously build runnable mentalmodels that can be used in problem solving so special illustra-tions are not needed.

Subjects and design. The subjects were 96 college students re-cruited from the psychology subject pool at the University of Califor-nia, Santa Barbara. Twenty-four students served in each of the fourtreatment groups: (1) the no illustrations group read a booklet aboutbraking systems that contained no illustrations; (2) the parts illustra-tions group read a booklet that contained illustrations showing themajor parts within each type of braking system; (3) the steps illustra-tions group read a booklet containing illustrations showing the majoractions occurring for each type of braking system; and (4) the parts-and-steps illustrations group received a booklet with both types ofillustrations. Half of the students in each group were low prior-knowledge students who rated their knowledge of automobile me-chanics as "very little" and reported that they never had performedany automobile repairs, whereas the other half were high prior-knowledge students who rated their knowledge of automobile me-chanics as more than "very little" and reported having performedminor car maintenance (such as changing oil or installing a sparkplug).

Materials. The materials consisted of four versions of a bookleton braking systems, a subject questionnaire, and three posttests, eachtyped on 8.5 x 11 inch sheets of paper.

The text in the booklet was taken from the World Book Encyclo-pedia's (1987) entry for "brakes," containing 750 words and 95 ideaunits. The text included an explanation of how each of four brakingsystems operate: mechanical braking (such as on bicycles), hydraulicdisk braking systems (such as on the front wheels of cars), hydraulicdrum braking systems (such as on the rear wheels of cars), and airbraking systems (such as on trains and trucks). For each type ofbraking system, the text included a description of how a change inthe status of one part affects changes in the status of other parts (i.e.,what Mayer, 1984, has called explanative information) as well asfactual information such as historical information or descriptions ofthe materials used to make the parts (i.e., what Mayer, 1984, hascalled nonexplanative information).

The no-illustrations version contained only the text; the parts-illustrations version added illustrations showing the status of each ofthe four braking systems before the brakes are applied, along withlabels for the major parts (using names repeated from the text); thesteps-illustrations version added illustrations showing the status ofeach of the four braking systems after the brakes are applied, alongwith a list of the major changes that occur (using wording repeatedfrom the text); the parts-and-steps illustrations version added boththe parts and the steps illustrations for each of the four brakingsystems. Illustrations appeared on the same page and below theircorresponding paragraphs. A portion of the text is given in Table 1and examples of each type of illustration are given in Figure 1.

The subject questionnaire solicited information concerning thestudents' experiences with automobile mechanics and repair. Forexample, students were asked to use a 5-point scale ranging from"very little" to "very much" for the item: "Please put a check markindicating your knowledge of car mechanics and repair." In anotheritem, students were asked: "Please place a check mark next to thethings you have done." The list included having a drivers' license,putting air into a tire, changing a tire, changing oil, installing sparkplugs, and replacing brake shoes. The recall posttest asked studentsto "write down all you can remember from the passage you have justread" and to "pretend that you are writing an encyclopedia forbeginners."

The problem-solving posttest consisted of five questions, eachprinted on a separate sheet: (a) Why do brakes get hot? (b) Whatcould be done to make brakes more reliable, that is, to make sure

WHAT MAKES A GOOD ILLUSTRATION? 719

Table 1Portion of Text on Brakes (With ExplanativeInformation in Italics)

Hydraulic brakes use various fluids instead of levers or cables. Inautomobiles, the brake fluid is in chambers called cylinders. Metaltubes connect the master cylinder with wheel cylinders located nearthe wheels. When the driver steps on the car's brake pedal, a pistonmoves forward inside the master cylinder. The piston forces brakefluid out of the master cylinder and through the tubes to the wheelcylinders. In the wheel cylinders, the increase in fluid pressure makesa set of smaller pistons move. These smaller pistons activate eitherdrum brakes or disk brakes. Most automobiles have drum brakes onthe rear wheels and disk brakes on the front wheels.

Drum brakes consist of a cast-iron drum and a pair of semicircularbrake shoes. The drum is bolted to the center of the wheel on theinside. The drum rotates with the wheel, but the shoes do not. Theshoes are lined with asbestos or some other material that can with-stand heat generated by friction. When the brake shoes press againstthe drum, both the drum and the wheel stop or slow down.

Note. Adapted from The World Book Encyclopedia (Vol. 2, p. 463),1987, Chicago: World Book, Inc. Copyright 1990 by World Book,Inc. by permission of the publisher.

they would not fail? (c) What could be done to make brakes moreeffective, that is, to reduce the distance needed to stop? (d) Supposethat you press on the brake pedal in your car but the brakes don'twork. What could have gone wrong? (e) What happens when youpump the brakes (i.e., press the pedal and release the pedal repeatedlyand rapidly)?

The verbatim retention posttest consisted of eight pairs of sen-tences. For each pair, one of the sentences had occurred verbatim inthe text whereas the other was a reworded version that retained theoriginal meaning. For example, one item on the verbatim recognitionposttest was: "Each car on a train has its own tank of compressed air.On trains, each car has its own tank of compressed air." The instruc-tions at the top of the sheet asked the student to "place a check marknext to the sentence that is exactly (word-for-word) identical to asentence in the passage you read."

Procedure. Students were randomly assigned to treatment groups.First, the student filled out the subject questionnaire. Second, thestudent was given 8 minutes to read a booklet corresponding to hisor her treatment group. Third, the student was allowed 10 minutesto take the recall posttest. Fourth, the student was allowed 12.5minutes to take the transfer posttest; under the experimenter's instruc-tions the subjects spent 2.S minutes on each item and were notallowed to go back to previous items. Finally, the students wereallowed 3 minutes to complete the verbatim retention posttest.

Results and Discussion

Scoring. Students who rated their mechanical experienceas "very little" and who indicated having had two or fewer ofthe mechanical experiences listed on the subject questionnairewere counted as low-knowledge students. Students who ratedtheir mechanical experience as more than "very little" andindicated having had more than two mechanical experienceswere counted as high-knowledge students.

For each student, four major dependent measured wererecorded: number of explanative idea units recalled (out of35 possible); number of nonexplanative idea units recalled(out of 60 possible); number of correct answers on the prob-lem-solving test (out of 15 possible); and number of correctanswers on the verbatim recognition test (out of 8 possible).

Each student's recall protocol was broken down into ideaunits; credit was given if an idea unit in the student's recallprotocol conveyed the same meaning or contained the samekey words as an idea unit in the passage (see Mayer, 1985,1989b). The verbatim retention test was scored by tallying thenumber of correct answers. The problem-solving test wasscored by tallying the number of correct answers producedfor each problem, as described by Mayer (1989b). Examplesof acceptable answers for the five problem-solving questionsare: (a) Heat is caused by friction, pressure, rubbing, orpressing a stationary object (such as a brake shoe) against arotating object (such as a rim or disk); (b) Reliability couldbe increased by adding a backup braking system, by usingthicker shoes, tougher tubes, or stronger cables, or by devel-oping more heat-resistant pads or shoes or a cooling system;(c) Effectiveness could be increased by injecting the brakefluid into the tubes or using faster moving fluid, by using amore friction-sensitive shoe, or by decreasing the distancebetween the shoe and the pad; (d) Brakes may malfunctionbecause of lack of fluid, worn or loose pads, a jammed pistonor holes in the piston, a hole in the tube or a break in thecable line, or moisture between pads and shoes; (e) Pumpingthe brakes reduces the build up of heat and allows the pad topress on the shoe at more than one place.

In analyzing the results, Predictions 1-6 (discussed previ-ously) were tested. Predictions 1 and 2 examined the types oftests that are indicators of effective illustrations; Predictions

Low PriDr-Knowlfitipe I earners

• Control0 Paris0 Steps• Parts & Steps PI

999999.99.99.

High Prior-Knowledge Learners• ControlE PansQ Steps• Parts & Steps

v v949494

94^2

ConceptualRecall

NonconceplualRecall

ProblemSolving

VerbatimRetention

Figure 4. Proportion correct on four posttests by treatment groupfor low and high prior-knowledge learners in Experiment 1.


3 and 4 assessed the types of illustrations that lead to improvedperformance; and Predictions 5 and 6 focused on how thetypes of learners relate to the effectiveness of illustrations.

Prediction 1: Explanative illustrations improve recall ofexplanative information but not nonexplanative recall. Thefirst prediction is that low prior-knowledge students who readexplanative text including explanative illustrations (i.e., parts-and-steps illustrations) will recall more explanative informa-tion but not more nonexplanative information as comparedto low prior-knowledge students who read text without illus-trations. The top-left portion of Figure 4 shows the proportioncorrect response for low prior-knowledge students in eachtreatment group on explanative and nonexplanative recall.As can be seen, the parts-and-steps group outperformed thecontrol group on recall of explanative information (35%versus 15%) but not on recall of nonexplanative information(19% versus 20%). Consistent with Prediction 1 and theresults of Mayer (1989b), separate t- tests conducted on thesedata revealed that the groups differed in recall of explanativeinformation, t{22) « 4.53, p < .001, but not in recall ofnonexplanative information, t < 1.

Prediction 2: Explanative illustrations improve creativeproblem solving but not verbatim retention. The second pre-diction is that low prior-knowledge students who read explan-ative text including explanative illustrations (i.e., both partsand steps) will generate more creative answers to transferproblems but will not perform better on verbatim retentionof sentence wording as compared to low prior-knowledgestudents who read the text without illustrations. The top-rightportion of Figure 4 shows the proportion correct response forlow prior-knowledge students in each treatment group onproblem solving and verbatim retention. As can be seen, theparts-and-steps group outperformed the control group onproblem solving (43% versus 24%) but not on verbatimretention (40% versus 45%). Consistent with Prediction 2 andthe results of Mayer (1989b), separate t- tests conducted onthese data revealed the groups differed in problem solvingperformance ;(22) = 4.17. p < .001, but not in verbatimretention, t < 1.

Prediction 3: Explanative illustrations are more effective inimproving conceptual recall than nonexplanative illustra-tions. The third prediction is that low prior-knowledge stu-dents who read explanative text containing explanative illus-trations (i.e., parts-and-steps) will show increases in explana-tive recall over the control group, but low prior-knowledgestudents who read explanative text with static illustrations(i.e., steps or parts) will not differ from the control group. Thetop-left portion of Figure 4 shows that the parts-and-stepsgroup outperformed the steps, parts, and control groups onrecall of explanative information (35%, 21%, 17%, and 15%,respectively). Consistent with Prediction 3, an analysis ofvariance conducted on these data revealed the groups differedin recall of explanative information, F(3, 44) - 9.81, MSC —.33, p < .001, and supplemental Dunnett's Tests (at p < .05)revealed that the parts-and-steps group outperformed thecontrol group, but the other illustrations groups did notoutperform the control group.

Prediction 4: Explanative illustrations are more effective inimproving problem solving performance than nonexplanative

illustrations. The fourth prediction is that low prior-knowl-edge students who read explanative text containing explana-tive illustrations (i.e., parts-and-steps) will show increases inproblem solving over the control group, but low prior-knowl-edge students who read explanative text with static illustra-tions (i.e., steps or parts) will not differ from the controlgroup. The top-right portion of Figure 4 shows that the parts-and-steps group outperformed the steps, parts, and controlgroups on problem-solving (43%, 26%, 26%, and 24%, re-spectively). Consistent with Prediction 4, an analysis of vari-ance conducted on these data revealed that the groups differedin problem-solving performance, F(3, 44) = 5.78, MSe =.31,p < .002, and supplemental Dunnett's Tests (at p < .05)revealed that the parts-and-steps group outperformed thecontrol group, but the other illustrations groups did notoutperform the control group.

Prediction 5: Explanative illustrations improve recall ofexplanative information for low-knowledge students but notfor high-knowledge students. The fifth prediction is that thestrong pattern of performance on explanative recall obtainedfor low-prior knowledge students (described under Prediction3) will be attenuated or eliminated for high-knowledge stu-dents. The bottom-left panel of Figure 4 shows the proportioncorrect response for high prior-knowledge students in eachtreatment group on explanative recall. As can be seen, forhigh prior-knowledge students, the parts-and-steps, steps,parts, and control groups do not seem to differ in explanativerecall (28%, 34%, 30%, and 32%, respectively). Consistentwith Prediction 5, an analysis of variance conducted on thesedata for high prior-knowledge students revealed no significantdifferences among the groups, F(3, 44) = .61, MSC = .52, ns\in contrast, as described under Prediction 3, the explanativeillustrations were effective in improving explanative recall forlow prior-knowledge students.

Prediction 6: Explanative illustrations improve creativeproblem solving for low-knowledge students but not for high-knowledge learners. The sixth prediction is that the strongpattern of performance on problem solving obtained for lowprior-knowledge students (described under Prediction 4) willbe attenuated or eliminated for high-knowledge students. Thebottom-right panel of Figure 4 shows the proportion correctresponse for high prior-knowledge students in each treatmentgroup on problem solving. As can be seen, for high prior-knowledge students, the parts-and-steps, steps, parts, and con-trol groups do not seem to differ in problem solving (45%,43%, 38%, and 39%, respectively). Consistent with Prediction6, an analysis of variance conducted on these data for highprior-knowledge students revealed no significant differencesamong the groups F(3, 44) = 1.17, MSe = .22, ns; in contrast,as described under Prediction 4, the explanative illustrationswere effective in improving problem solving for low prior-knowledge students.

Experiment 2 (Pumps)

Experiment 2 used a passage on how pumps work and wasintended to provide replicative tests of the six predictions.


Method

Subjects and design. The subjects were 96 college students re-cruited from the same subject pool as in Experiment 1. The designand cell sizes were identical to Experiment 1, except that high andlow prior-knowledge status was based on students' experience withhousehold repair rather than car mechanics.

Materials. Similar to Experiment 1, the materials consisted offour versions of a booklet on pumps, a subject questionnaire, andthree posttests, each typed on 8.5 x 11 inch sheets of paper, The textwas taken from the World Book Encyclopedia's (1987) entry for"pump," containing 812 words and 128 idea units. The text includedexplanations of how centrifugal, sliding vane, lift, and bicycle tirepumps operate. No illustrations, parts illustrations, steps illustrations,and parts-and-steps illustrations booklets were constructed as in Ex-periment 1. A portion of the text is given in Table 2 and examples ofeach type of illustration are given in Figure 2.

The subject questionnaire was similar to that used in Experiment1 except that it solicited information concerning the students' expe-rience with household repair rather than automobile mechanics. Forexample, students were asked to use a 5-point scale ranging from"very little" to "very much" for the item: "Please put a check markindicating your knowledge of how to fix household appliances andmachines." In another item, students were asked: "Please place acheck mark next to the things you have done." The list includedowning a screw driver, owning a power saw, having replaced theheads on a lawn sprinkler system, having replaced the washer in asink faucet, having replaced the flush mechanism in a toilet, andhaving installed plumbing pipes or fixtures. The recall posttest wasidentical to the one used in Experiment 1.

As in Experiment 1, the problem-solving posttest consisted of fivequestions, each on a separate sheet. The five questions were: (a) Whatcould be done to make a pump more reliable, that is, to make sure itwould not fail? (b) What could be done to make a pump moreeffective, that is, to move more liquid or gas more rapidly? (c) Supposeyou push down and pull up the handle of a lift pump several timesbut no water comes out. What could have gone wrong? (d) Why doeswater enter a lift pump? Why does water exit from a lift pump? (e)The text you read mentioned a "screw pump that consisted of a screwrotating in a cylinder," but the text did not really explain how itworks. Based on your understanding of how pumps work, pleasewrite your own idea of how you think a screw pump could be usedto move water.

The verbatim retention posttest corresponded to that used inExperiment 1 except that the eight sentence pairs were based on thepumps text. For example, one item on the verbatim recognition

Table 2Portion of Text on Pumps (With ExplanativeInformation in Italics)

Bicycle tire pumps vary in the number and location of the valvesthey have and in the way air enters the cylinder. Some simple bicycletire pumps have the inlet valve on the piston and the outlet valve atthe closed end of the cylinder. A bicycle tire pump has a piston thatmoves up and down. Air enters the pump near the point where theconnecting rod passes through the cylinder. As the rod is pulled out,air passes through the piston and fills the areas between the pistonand the outlet valve. As the rod is pushed in, the inlet valve closes andthe piston forces air through the outlet valve.

Note. Adapted from The World Book Encyclopedia (Vol. 15, p.794), 1987, Chicago: World Book, Inc. Copyright 1990 by WorldBook, Inc. by permission of the publisher.

posttest was: "Most pumps are made of steel, but some are made ofglass or plastic. Usually, pumps are made of steel, but sometimes theyare made of glass or plastic."

Procedure. The procedure was identical to that used in Experi-ment 1 except that students had 10 minutes to read their booklets.


Scoring. Students who rated their knowledge of householdrepair as "very little" and who indicated having had two orfewer of the experiences listed on the subject questionnairewere counted as low prior-knowledge students; students whorated their household-repair knowledge as more than "verylittle" and indicated having had more than two household-repair experiences were counted as high prior-knowledge stu-dents.

As in Experiment 1, four major dependent measures wererecorded for each subject: number of explanative idea unitsrecalled (out of 41 possible); number of nonexplanative ideaunits recalled (out of 87 possible); number of correct answerson the problem-solving test (out of 15 possible); and numberof correct answers on the verbatim recognition test (out of 8possible). The recall, problem-solving, and verbatim retentiontests were scored as in Experiment 1. Examples of acceptableanswers for the five problem-solving questions are: (a) Relia-bility could be increased by using airtight seals on valves, byusing a backup system, and by using roller bearings to helpthe impeller turn more easily; (b) Effectiveness could beincreased by turning the impeller faster, using larger intakeand output pipes, and increasing the diameter of the cylinder;(c) A lift pump might malfunction because there is no waterin the pump, a valve is stuck, a seal is broken, a blockage hasformed in the line, and the piston has become unattachedfrom the handle; (d) Water enters a lift pump because ofsuction or a vacuum, whereas water exits because of compres-sion or pressure, and (3) A screw pump has a rotating screwwith threads that seals tightly against the inside of the cylinder.

In analyzing the results, we tested the same six predictionsas in Experiment 1.

Prediction I: Explanative illustrations improve recall ofexplanative information but not nonexplanative recall. Thetop-left portion of Figure 5 shows the proportion correctresponse for low prior-knowledge students in each treatmentgroup on explanative and nonexplanative recall. As can beseen, the parts-and-steps group outperformed the controlgroup on recall of explanative information (23% versus 3%)but not on recall of nonexplanative information (9% versus16%). Consistent with Prediction 1, the results of ExperimentI, and the results of Mayer (1989b), separate t- tests conductedon these data revealed that the parts-and-steps group outper-formed the control group in recall of explanative information,t(28) = 4.76, p < .001. In addition, the parts-and-steps groupnot only failed to outperform the control group on recall ofnonexplanative information as in Experiment 1; in Experi-ment 2, the control group actually recalled significantly morenonexplanative information than the control group, Z(28) =4.19, p < .001, as would be predicted by a strong version ofschema theory.


u ,5 -

o .3-

.2-

Low Prior-Knowledge Learners

• ControlE2 ParisQ StepsH Parts & Steps

I

99999%9 v9 V9 09 y9 v9999999i9i9 V9 v9 y9 y9

ConcapiualRecall

NonconceptualRecall

ProblemSolving

u

.6-

.5-

.4-

o .3-

.2-

. 1 •

High Prior-Knowledge Learners

Q Control0 Parts0 Steps• Pans & Steps

VerbatimReleniion

J%94909 v9 v9%909 y9 v9 v9 099 >9%9v99

ConceptualRecall

NonconceptualRecall

ProblemSolving

VerbalimRetention


Prediction 2: Explanative illustrations improve creativeproblem solving but not verbatim retention. The top-rightportion of Figure 5 shows the proportion correct response forlow prior-knowledge students in each treatment group onproblem solving and verbatim retention. As can be seen, theparts-and-steps group outperformed the control group onproblem solving (47% versus 28%) but not on verbatimretention (66% versus 68%). Consistent with Prediction 2, theresults of Experiment 1, and the results of Mayer (1989b),separate t- tests conducted on these data revealed that thegroups differed in problem solving performance, /(28) = 3.51,p < .01, but not in verbatim retention, /(28) < 1.

Prediction 3: Explanative illustrations are more effective inimproving conceptual recall than nonexplanative illustra-tions. The top-left portion of Figure 5 shows that the parts-and-steps group outperformed the steps, parts, and controlgroups on recall of explanative information (23%, 13%, 13%,and 3%, respectively). Consistent with Prediction 3 and theresults of Experiment 1, an analysis of variance conducted on

these data revealed the groups differed in recall of explanativeinformation, F(3, 56) = 7.32, MSe = .53, p < .001, andsupplemental Dunnetf s Tests (at p < .05) revealed that theparts-and-steps group outperformed the control group, butthe other illustrations groups did not outperform the controlgroup.

Prediction 4: Explanative illustrations are more effective inimproving problem solving performance than nonexplanativeillustrations. The top-right portion of Figure 5 shows thatthe parts-and-steps group outperformed the steps, parts, andcontrol groups on problem solving (47%, 20%, 28%, and28%, respectively). Consistent with Prediction 4 and the re-sults of Experiment 1, an analysis of variance conducted onthese data revealed the groups differed in problem-solvingperformance, F(3, 56) = 9.20, MSe = .31, p < .001, andsupplemental Dunnett's Tests (at p < .05) revealed that theparts-and-steps group outperformed the control group, butthe other illustrations groups did not outperform the controlgroup.

Prediction 5: Explanative illustrations improve recall ofexplanative information for low-knowledge students but notfor h igh-knowledge students. The bottom-left panel of Figure5 shows the proportion correct response for high prior-knowl-edge students in each treatment group on explanative recall.As can be seen, for high prior-knowledge students, the differ-ences among the parts-and-steps, steps, parts, and controlgroups in explanative recall (24%, 23%, 16%, and 13%,respectively) were relatively small. However, in contrast toPrediction 5 and the results of Experiment 1, an analysis ofvariance conducted on these data for high prior-knowledgestudents revealed mildly significant differences among thegroups, F(3, 56) = 2.98, MSC = .53, p < .05. Apparently, theemphasis on explanative information in the illustrations in-fluenced the recall performance of high prior-knowledge inthis experiment,

Prediction 6: Explanative illustrations improve creativeproblem solving for low-knowledge illustrations but not forhigh know/edge learners. The bottom-right panel of Figure5 shows the proportion correct response for high prior-knowl-edge students in each treatment group on problem solving.As can be seen, for high prior-knowledge students, the parts-and-steps, steps, parts, and control groups do not seem todiffer in problem solving (55%, 45%, 49%, and 51%, respec-tively). Consistent with Prediction 6, an analysis of varianceconducted on these data for high prior-knowledge studentsrevealed no significant differences among the groups, F(3, 56)- 5.66, MSC = .72, ns\ in contrast, as described under Predic-tion 4, the explanative illustrations were effective in improv-ing problem solving for low prior-knowledge students.

Experiment 3 (Generators)

Experiment 3 used a passage on electric generators and wasintended to provide replicative tests of the six predictions.This experiment differed from Experiments 1 and 2 in severalimportant ways: the passage was longer and more complex,there were only three versions of the booklet rather than four,the parts-and-steps illustrations contained labels for parts butnot for steps, the parts-and-steps illustrations booklet included


several illustrations that were not in the same format as usedin previous experiments, words were added to the parts andparts-and-steps booklets to clarify the illustrations, and priorknowledge was measured on a different kind of survey instru-ment.

Method

Subjects and design. The subjects were 108 college students re-cruited from the same subject pool as Experiments 1 and 2. Thedesign was identical to Experiment 1 except that there was no stepsgroup; the no-illustrations group consisted of 17 low prior-knowledgeand 18 high prior-knowledge students; the parts group consisted of15 low prior-knowledge and 21 high-prior knowledge students; andthe parts-and-steps group consisted of 12 low-prior knowledge and25 high prior-knowledge students.

Materials. The materials consisted of three versions of a passageon electric generators, a subject questionnaire, and three posttests,2

each typed on 8.5 x 11 in sheets of paper.The text for the no-illustrations booklet was taken from the World

Book Encyclopedia's (1987) entry on "electric generator," containing2066 words and 161 idea units. The parts-illustrations booklet wascreated by inserting 4 figures and approximately 225 words thatclarified the figures. The figures showed the parts in a simplifiedgenerator, a real-life generator, a simplified AC generator, and a real-life AC generator. The parts-and-steps illustrations booklet was cre-ated by adding 13 figures and 372 words that clarified the figures. Itincluded the same figures as in the parts booklet along with thefollowing additions: (a) three additional illustrations of the simplegenerator showing the armature in each state of rotation; (b) threeadditional illustrations of the simplified AC generator showing thearmature in each state of rotation; (c) two additional illustrationsshowing the relation between the simplified and real-life generatorand between the simplified AC generator and real-life AC generator;and (d) a summary illustration showing the flow of current amongthe parts in a simplified AC generator.

The subject questionnaire consisted of 11 statements concerningthe subject's experience and interest in working with mechanical andelectrical devices. Each statement was to be rated on a 5-point scaleconsisting of never (1), seldom (2), sometimes OX frequently (4), andalways (5). Typical questionnaire items included: "I like to repairelectrical gadgets." "I enjoy experimenting with mechanical things,"and "I like to disassemble mechanical things merely to perceive whatthey look like inside."

The recall posttest was identical to that used in Experiments 1and 2.

The problem-solving posttest contained eight questions: (a) Whyis no energy generated when the armature of the generator is in aparallel position? (b) Provide reasons as to why a generator might notproduce enough electrical energy, (c) What can be done to increasethe energy output from an electrical generator? (d) How would youknow if an AC electric generator is not functioning properly? (e)Explain why one generator might produce more cycles of currentthan another? (0 What happens to the electric current if no device ishooked up to the electric generator? (g) Can an AC generator everconduct electric current in the same direction throughout one com-plete cycle of the rotation process? Explain your answer, (h) Supposea radio is hooked up to an AC generator. The radio suddenly goesout. The problem seems to be related to the generator. Provide anexplanation for the power failure.

The verbatim retention posttest consisted of 15 pairs of sentencessuch as, "An electric generator is a machine that produces electricity.An electric generator is a machine that generates electricity.

Procedure. The procedure was the same as in Experiments 1 and2 except that subjects were given 25 minutes to read their booklets.


Scoring. Students' scores on the subject questionnairewere computed by tallying the numbers (i.e., 1 through 5)circled for each of the eleven items. Students who obtainedscores below 29 were counted as low prior-knowledge studentsand those who obtained scores of 29 or above were countedas high prior-knowledge students.

As in Experiments 1 and 2, four major dependent measureswere recorded for each subject: number of explanative ideaunits recalled (out of 32 possible); number of nonexplanativeidea units recalled (out of 129 possible); number of correctanswers on the problem-solving test (out of 15 possible); andnumber of correct answers on the verbatim recognition test(out of 15 possible). The recall, problem-solving, and verbatimretention tests were scored as in Experiment 1. Examples ofacceptable answers for the problem-solving questions are: (a)No energy is generated when the armature is in a parallelposition because it is not cutting through the magnetic linesof force; (b) The generator might not produce enough electri-cal energy because the carbon brushes are not properly alignedwith the slip ring, or the initial current from the prime moveris insufficient; (c) One sign of an AC generator malfunctioningis that the current does not reverse direction; (d) A generatormight produce more cycles of current than another becauseit is larger or the armature is rotating at a faster rate; (e) If nodevice is hooked up to an electric generator, no electricalcurrent is produced; (f) A generator that conducts electricalcurrent in the same direction is not an AC generator but aDC generator; and (g) If a radio hooked up to a generatorsuddenly fails, the problem could be that the armature stoppedrotating or the slip rings are not rotating properly.

As in Experiments 1 and 2, six predictions were tested.Prediction 1: Explanative illustrations improve recall of

explanative information but not nonexplanative recall. Thetop-left portion of Figure 6 shows the proportion correctresponse for low prior-knowledge students in each treatmentgroup on explanative and nonexplanative recall. As can beseen, the parts-and-steps group outperformed the controlgroup on recall of explanative information (21% versus 9%)but not on recall of nonexplanative information (7% versus8%). Consistent with Prediction 1, the results of Experiments1 and 2, and the results of Mayer (1989b), separate t-testsconducted on these data revealed that the groups differed inrecall of explanative information, t(27) = 3.37, p < .01, butnot in recall of nonexplanative information, /(27) < 1, ns.

Prediction 2: Explanative illustrations improve creativeproblem solving but not verbatim retention. The top-rightportion of Figure 6 shows the proportion correct response forlow prior-knowledge students in each treatment group onproblem solving and verbatim retention. As can be seen, theparts-and-steps group outperformed the control group on

2 Several additional posttests were given but were not analyzedbecause they were not relevant to our six predictions.


S -3-

.2 -

UQW Pripr-Knowledge Learners

D Control0 Paris| Parts & Steps

ConceptualRecall

NonconceptjalRecall

ProblemSolving

VerbatimRetention

High Prior-Knowledge Learners

Q ControlE3 Parts• Paris & Sieps

ConceptualRecall

NonconceplualRecall

ProblemSolving

VerbatimRetention


problem solving (56% versus 29%) but not on verbatimretention (67% versus 75%). Consistent with Prediction 2, theresults of Experiments 1 and 2, and the results of Mayer(1989b), separate t- tests conducted on these data revealedthat the groups differed in problem-solving performance t(21)= 4.59, p < .001, but not in verbatim retention, r(27) = 1.32,ns.

Prediction 3: Exploitative illustrations are more effective inimproving conceptual recall than nonexplanative illustra-tions. The top-left portion of Figure 6 shows that the parts-and-steps group outperformed the parts and control groupson recall of explanative information (21%, 12%, and 9%,respectively). Consistent with Prediction 3, an analysis ofvariance conducted on these data revealed that the groupsdiffered in recall of explanative information, F(2, 40) =11.348, p < .001, and supplemental Dunnett's Tests (at p <.05) revealed that the parts-and-steps group outperformed thecontrol group, but the parts groups did not outperform thecontrol group.

Prediction 4: Explanative illustrations are more effective inimproving problem solving performance than nonexplanativeillustrations. The top-right portion of Figure 6 shows thatthe parts-and-steps group outperformed the parts and controlgroups on problem solving (56%, 39%, and 29%, respec-tively). Consistent with Prediction 4, an analysis of varianceconducted on these data revealed that the groups differed inproblem-solving performance F(2, 40) = 21.1 \,p< .001,andsupplemental Dunnett's Tests (at p < .05) revealed that theparts-and-steps group outperformed the control group, butthe parts group did not outperform the control group.

Prediction 5: Explanative illustrations improve recall ofexplanative information for low-knowledge students but notfor high-knowledge students. The bottom-left panel of Figure6 shows the proportion correct response for high prior-knowl-edge students in each treatment group on explanative recall.As can be seen, for high prior-knowledge students, the differ-ences among the parts-and-steps, parts, and control groups inexplanative recall (26%, 16%, and 11%, respectively) wererelatively large. Inconsistent with Prediction 5 and the resultsof Experiments 1 and 2, an analysis of variance conducted onthese data for high prior-knowledge students showed signifi-cant differences among the groups, F(2, 60) = 19.96, p <.001, and is similar to the pattern for low prior-knowledgestudents described under Prediction 3 in which explanativeillustrations were effective in improving explanative recall.Thus, both in Experiments 2 and 3, this is the only inconsist-ency in our predictions.

Prediction 6: Explanative illustrations improve creativeproblem solving for low-knowledge students but not for high-knowledge learners. The bottom-right panel of Figure 6shows the proportion correct response for high prior-knowl-edge students in each treatment group on problem solving.As can be seen, for high prior-knowledge students, the parts-and-steps, parts, and control groups do not seem to differ inproblem solving (61%, 55%, and 55%, respectively). Consist-ent with Prediction 6 and the results of Experiments 1 and 2,an analysis of variance conducted on these data for high prior-knowledge students revealed no significant differences amongthe groups, F(2,60) = 1.20, ns; in contrast, as described underPrediction 4, the explanative illustrations were effective inimproving problem solving for low prior-knowledge students.

Conclusion

Conditions for Effective Illustrations

A general goal of this study was to test the model ofconditions for effective illustrations summarized in Figure 3.We examined the specific predictions that illustrations wouldbe effective in our experiments only with explanative text(i.e., appropriate text), tests of understanding and reasoning(i.e., appropriate tests), steps-and-parts illustrations (i.e., ap-propriate illustrations), and low prior-knowledge learners (i.e.,appropriate learners). Table 3 summarizes whether or noteach of the six predictions, based on the model of conditionsfor effective illustrations, was supported in each of the threeexperiments.


Table 3Summary of Results: Tests of18 Predictions

Predictions based on four conditions for effective illustrations (in boldface)

Results of Experiment

Appropriate text: Use exploitative text(No predictions were tested.)Appropriate test: Use tests that measure conceptual understandingExplanative illustrations improve explanative recall but not nonexplanative

recall. Y Y YExplanative illustrations improve creative problem solving but not verbatim

retention. Y Y YAppropriate illustrations: Use illustrations that explainExplanative illustrations improve explanative recall better than nonexplanative

illustrations. Y Y YExplanative illustrations improve problem solving better than nonexplanative

illustrations. Y Y YAppropriate learners: Use illustrations for learners who lack prior knowledgeExplanative illustrations improve explanative recall for low rather than high

knowledge learners. Y N NExplanative illustrations improve problem-solving for low rather than high

knowledge learners. Y Y Y

Note. Y = prediction was confirmed', N = prediction was not confirmed.

Appropriate text. The first condition in the model is thatthe text be appropriate for the instructional goal. Since ourinstructional goal was to enhance learner understanding, weused explanative text rather than descriptive or narrative text.Although we did not directly manipulate the type of text inthis study, we exclusively used text in each experiment thatmet the condition of appropriateness.

Appropriate tests. The second condition is that the test beappropriate to the instructional goal. In our experiments weevaluated student performance on appropriate tests, such asexplanative recall and problem solving, and on inappropriatetests, such as nonexplanative recall and verbatim retention.The first two predictions in Table 3, relating to test appropri-ateness, were supported in each of the three experiments. Theconsistency of these results, in conjunction with similar resultsin two previous experiments (Mayer, 1989b), provide strongsupport for the idea that tests should be sensitive to the specificgoals of instruction.

Appropriate illustrations. The third condition is that theillustrations should be appropriate for the instructional goal.In our experiments, we compared illustrations that concre-tized the changes in status of parts within the system (explan-ative illustrations) to illustrations that did not (nonexplanativeillustrations). The second two predictions in Table 3, concern-ing illustration appropriateness, were supported in each of thethree experiments. These results provide new evidence con-cerning the characteristics of the models portrayed in effectiveexplanative illustrations: The ineffective illustrations failed tovisually portray either system topology (i.e., steps illustrations)or component behavior (i.e., parts illustrations) whereas theeffective illustrations (i.e., parts-and-steps illustrations) por-trayed both. In particular, the parts-and-steps illustrationsused a series of two (or more) frames to show the state of thecomponents within the system at various points in the oper-ation of the system.

Appropriate learners. The fourth condition is that thelearners would not achieve the desired learning outcome

without special instruction. In our experiments, we comparedthe learning outcomes of students who lacked domain-specificknowledge and those who possessed it. The final two predic-tions in Table 3, testing learner appropriateness, were gener-ally supported. The major exception, that explanative illiistra-tions improved explanative recall for high-knowledge learnersin Experiment 3, may be accounted for by the fact that theexplanative text was expanded upon in Experiment 3 but notin the other experiments.

In summary, in the spirit of complementing Larkin &Simon's (1987) work, our goal was to answer the question,"When is a diagram worth ten thousand words?" Our resultsprovide a four-part answer: when the text is potentially un-derstandable, when the value of illustrations is measured interms of learner understanding, when the illustrations explain,and when the student lacks previous experience. Finally, ourreport fits into a growing literature on the cognitive psychol-ogy of visual instructional methods (Holley & Dansereau,1984; Waddill, McDaniel & Einstein, 1988; Weidenmann,1989; Winn, 1987) and points to the potential of visuallybased instruction as a medium for promoting students' un-derstanding of scientific material.

References

Alexander, P. A., & Judy, J. E. (1988). The interaction of domain-specific and strategic knowledge in academic performance. Reviewof Educational Research, 58, 375-404.

Bobrow, D. G. (Ed.). (1985). Qualitative reasoning about physicalsystems. Cambridge, MA: MIT Press.

de Kleer, J., & Brown, J. S. (1983). Assumptions and ambiguities inmechanistic mental models. In D. Gentner & A. L. Stevens (Eds.),Mental models (pp. 172-190). Hillsdale, NJ: Erlbaum.

de Kleer, J., &. Brown, J. S. (1985). A qualitative physics based onconfluences. In D. G. Bobrow (Ed.), Qualitative reasoning aboutphysical systems (pp. 7-83). Cambridge, MA: MIT Press.

Gentner, D., & Gentner, D. R. (1983). Flowing waters or teemingcrowds: Mental models of electricity. In D. Gentner & A. L. Stevens


(Eds.), Mental models (pp. 101-130). Hillsdale, NJ: Erlbaum.Gentner, D., & Stevens, A. L. (Eds.). (1983). Mental models. Hillsdale,

NJ: Erlbaum.Holley, C. D., & Dansereau? D. F. (Eds.). (1984). Spatial learning

strategies. San Diego, CA: Academic Press.Kieras, D. E., & Bovair, S. (1984). The role of a mental model in

learning to operate a device. Cognitive Science, 8, 255-274.Larkin, J. H., & Simon, H. A. (1987). Why a diagram is (sometimes)

worth ten thousand words. Cognitive Science, 11, 65-99.Levie, W. H., Sc Lentz, R. (1982). Effects of text illustrations: A

review of research. Educational Communication and TechnologyJournal, SO, 195-232.

Levin, J. R. (1981). On the functions of pictures in prose. In F. J.Pirozzolo & M. C. Wittrock (Eds.), Neuropsychological and cogni-tive processes in reading {pp. 203-228). San Diego, CA: AcademicPress.

Levin, J. R., Anglin, G. J., & Carney, R. N. (1987). On empiricallyvalidating functions of pictures in prose. In D. M. Willows Sc H.A. Houghton (Eds.), The psychology of illustration: Vol. 1. Basicresearch (pp. 51-85). New York: Springer-Verlag.

Mandl, H., & Levin, J. R. (Eds.). (1989). Knowledge acquisition fromtext and pictures. Amsterdam: Elsevier.

Mayer, R. E. (1984). Aids to text comprehension. Educational Psy-chologist, 19, 30-42.

Mayer, R. E. (1985). How to analyze science prose. In B. K. Britton& J. B. Black (Eds.), Understanding expository text (pp. 305-313).Hillsdale, NJ: Erlbaum.

Mayer, R. E. (1989a). Models for understanding. Review of Educa-tional Research, 59, 43-64.

Mayer, R. E. (1989b). Systematic thinking fostered by illustrations inscientific text. Journal of Educational Psychology, 81, 240-246.

McDaniel, M. A., & Pressley, M. (1987). Imagery and related mne-monic processes. New York: Springer-Verlag.

Rumelhart, D. E., & Ortony, A. (1977). The representation of knowl-edge in memory. In R., C. Anderson, R. J. Spiro, & W. E. Montague(Eds.), Schooling and the acquisition of knowledge (pp. 99-135).Hillsdale, NJ: Erlbaum.

Waddill, P. J., McDaniel, M. A., & Einstein, G. O. (1988). Illustra-tions as adjuncts to prose: A text-appropriate processing approach.Journal of Educational Psychology, 80, 457-464.

Weidenmann, B. (1989). When good pictures fail: An information-processing approach to the effects of illustrations. In H. Mandl &J. R. Levin (Eds.), Knowledge acquisition from text and pictures(pp. 157-171). Amsterdam: Elsevier.

White, B. Y., & Frederiksen, J. R. (1987). Qualitative models andintelligent learning environments. In R. W. Lawler & M. Yazdani(Eds.), Artificial intelligence and education: Vol. 1. Learning envi-ronments and tutoring systems (pp. 281-305). Norwood, NJ: Ablex.

Willows, D. M., & Houghton, H. A. (Eds.). (1987). The psychologyof illustration: Vol. 1. Basic research. New York: Springer-Verlag.

Winn, B. (1987). Charts, graphs, and diagrams in educational mate-rials. In D. M. Willows & H. A. Houghton (Eds.), The psychologyof illustration: Vol. 1. Basic research (pp. 152-198). New York:Springer-Verlag.

The world book encyclopedia. (1987). Chicago: World Book.

Received September 11, 1989Revision received May 11, 1990

Accepted May 29, 1990

when is an illustration worth ten thousand...

Documents