document resume ed 335 022 ir 015 189 author wilson, … · 2013-08-02 · document resume ed 335...

DOCUMENT RESUME

ED 335 022 IR 015 189

AUTHOR Wilson, Brent; Cole, PeggyTITLA Cognitive Apprenticeships: An Instructional Design

Review.PUB DATE 91

NOTE 13p.; In: Proceedings of Selected ResearchPresentations at the Annual Convention of theAssocia,..ton for Educational Communications andTechnology; see IR 015 132.

PUB TYPE Speeches/Conference Papers (150)

EDRS PRICE MF01/PC01 Plus Postage.DESCRIPTORS Classification; *Cognitive Psychology; Comparative

Analysis; Computer Assisted Instruction; ElementarySecondary Education; Epistemology; *InstructionalDesign; Interactive Video; *Learning Theories;*Models; Postsecondary Education

IDENTIFIERS *Cognitive Apprenticeships

ABSTRACTThis discussion of the relationship between two

related disciplines--cognitive psychology and instructional design(ID)--characterizes instructional design as a more applieddiscipline, which concerns itself more with prescriptions and modelsfor designing instruction, while instructional psychologists conductempirical research on learning and instructional processes. It isposited that a problem-solving orientation to education is needed ifschoo]s are to achieve substantial learning outcomes, and the conceptof cognitive apprenticeships, which emphasize returning instructionto settings where worthwhile problems can be worked with and solved,is proposed as a possible solution to this problem. A brief review ofID models focuses on instructional-design theory, component displaytheory, and elaboration theory, pointing out that some of theteaching models proposed by cognitive researchers bear strongresemblance to traditional ID models. The design elements of thecognitive apprenticeship model are then reviewed and related totraditional ID concepts. Examples of cognitive apprenticeships aregiven for teaching writing, reading, math, and weather forecasting.It is conclud2d that, even th.:Jugh instructional-design theorists maychafe at the continuing need to revise their theories in light ofadvances in psychological theory, it is good for both fields for thedialogue to continue. (46 references) (BBM)

***********************************************************************Reproductions supplied by EDRS are the best that can be made

from the original document.***********************************************************************

tr-bo

Lin

0

U.S. DEPARTMENT OF EDUCATIONOffice of Educational Research and Improvement

EDUCATIONAL RESOURCES INFORMATIONCENTER (ERIC)

r This document has beeir reproduced asreceived from Ihe person or organizationorigintiting / 1

r Minor changes have been made to improvereproduction quality

Points of view or opinions stated in this doCumerit do not necessarily represent officialOEM position or policy

Mk:

Cognitive Apprenticeship& An Instructional Design Review

Authors:

Brent WilsonPeggy Cole

BEST COPY MILANI,

"PERMISSION TO REPRODUCE THISMATERIAL HAS BEEN GRANTED BY

Michael R. Simonson

TO THE EDUCATIONAL RESOURCESINFORMATION CEN't ER (ERIC)."

COGNITIVE APPRENTICESHIPS:AN INSTRUCTIONAL DESIGN REVIEWBrent G. Wilson and Peggy ColeUniversity of Colorado at Denver

MIThe field of instructional design (ID)

emerged more than 30 years ago as p- ycholo-gists and educators searched fo. effet.civemeans of planning and producing instruc-tional systems (Reiser, 1987; Merrill, Kowallis,ta Wilson, 1981). Since that time, instructionaldesigners have became more clearly differen-tiated from instructional psychologists workingwithin a cognitivist tradition (Resnick, 1981;Glaser, 1982; Glaser ea Basik, 1989). ID theo-rists tend to concern them& 'ves withprescriptions and models fo designinginstruction while instructional psychologistsconduct empirical research on learning andinstructional processes.

Of course, the distinction between design-ers and psychologists is never clear-cut. Overthe years, many psychologists have putconsiderable energy into the design andimplementation of experimental instructionalprograms. Because the two fields supportdifferent literature and theory bases, commu-nication between the two fields is oftenstrained. Thus much design work of cognitivepsychologists has gone relatively unnoticed byinstructional design theorists.

Collins, Brown, and colleagues (e.g., Collins,Brown, dr Newman, 1989) have developed aninstructional model derived from an analysisof the way apprentices work under experts intraditional sodeties and from the way peopleseem to learn in everyday informal environ-ments (Rogoff lc Lave, 1984); they have calledtheir model cognitive apprenticeships, andhave identified a list of features found in"ideal" learning environments. Instructionalstrategies, according to the Collins-Brownmodel, would include modeling, coaching,scaffolding and fading, reflection, and explo-ration. Additional strategies are offered forrepresenting content, for sequencing, and formaximizing benefits from social interaction.

Of course, many of Collins' recommendedstrategies resemble strategies found in theinstructional-design literature (e.g., Reigeluth,1983). Clearly both fields could benefit fromimproved communication concerningresearch findings and lessons learned frompractical tryout. With that goal in mind, thepurpose of this paper is to analyze the Collins-Brown cognitive apprenticeship model from

an instructional-design point of view. In addi-tion to general strategies and recommendedcomponents, several teaching systemsemploying cognitive apprenticeship ideals aredescribed. The resultant review should provevaluable in two ways: (1) cognitive psycholo-gists should be able to make a better corre-spondence between their models and currentID theory, hopefully seeing areas needingimprovement, and (2) instructional-designtheorists also should be able to see correspon-dences and differences, which may lead torevision or expansion of their current models.

The Need for Cognitive ApprenticeshipsThe cognitive apprenticeship model rests

on a somewhat romantic conception of the"ideal" apprenticeship into a complex domain(Brown, Collins, dr Duguid, 1989). In contrastto the classroom context, which tends toremove knowledge from its sphere of use,Collins et al. recommend returning instructionto settings where worthwhile problems can beworked with and solved. The need for a prob-lem-solving orientation to education is appar-ent from the difficulty schools are having inachieving substantial learning outcomes(Resnick, 1989).

Another way to think about the concept cfapprenticeship is Gott's (1988) notion of thelost apprenticeship," a growing problem inindustrial and military settings. She noted theeffects of the increased complexity andautomation of production systems. First, theneed is growing for high levels of expertise insupervising and using automated worksystems; correspondingly, the need for entrylevels of expertise is declining. Workers onthe job are more and more expected to beflexible problem solvers; human interventionis often most needed at points of breakdownor malfunction. At these points, the expert iscalled in. Experts, however narrow the domain,do more than apply canned job aids or trou-bleshooting algorithms; rather, they internalizeconsiderable knowledge which they can use tosolve flexibly problems in real time.

3

946

Gott's second observation relates to train-ing opportunities. Now, at a time when moreproblem-solving expertise is needed due tothe complexity of systems, fewer on-the-jobtraining opportunities exist for entry-levelworkers. There is often little or no chance forbeginning workers to acclimatize themselvesto the job, and workers very quickly areexpected to perform like seasoned profes-sionals. True apprenticeship experiences arebecoming relatively rare. Gott calls thisdilemmamore complex job requirementswith less time on the job to learnthe "lost"apprenticeship, and argues for the criticalneed for cognitive apprenticeships andsimulation-type training to help workersdevelop greater problem-solving expertise.

A Brief Review of ID ModelsCurrent instructional-design models are

based on Robert Gagne's conditions-of-learning paradigm (Gagne, 1985), which in itstime was a significant departure from theSkinnerian operant conditioning paradigmdominant among American psychologists.The conditions-of-learning paradigm positsthat a graded hierarchy of learning outcomesexists, and for each desired outcome, a set ofconditions exists that leads to learning.Instructional design is a matter of being clearabout intended learning outcomes, thenmatching up appropriate instructional strate-gies. The designer writes behaviorally specificlearning objectives, classifies those objectivesaccording to a taxonomy of learning types,then arranges the instructional conditions tofit the current instructional prescriptions. Inthis way, designers can design instruction tosuccessfully teach a rule, a psychomotor skill,an attitude, or piece of verbal information.

A related idea within the conditions-of-learning paradigm claims that sequencing ofinstruction should be based on a hierarchicalprogression from simple to complex learningoutcomes. R. Gagne developed a technique oflearning hierarchies for analyzing skills: A skillis rationally decomposed into parts and sub-parts, then instruction is ordered from simplesubskills to the complete skill. Elaborationtheory uses content structure (concept,procedure, or pAnciple) as the basis fororganizing and sequencing instruction(Reigeluth, Merrill, Wilson, & Spiller, 1980).Both depend on task analysis to break down

Cognitive Apprenticeship 3

the goals of instruction, then on a method ofsequencing proceeding from simple to grad-ually more complex and complete tasks.

These instructional-design models appearto work with well-defined content domains;their most extensive use has been in militarysettings. There are a couple of potentialproblems, however, that need to be addressed.First, sequencing of instruction based on thelogical structure of content tends to neglectthe existing knowledge base of individuallearners. ID models often assume newcontent can be added incrementally to learn-ers' prior knowledge bases without complka-Hon. As a simple example, imagine a concept"kinds" hierarchy of animals, vertebrates,mammals, dogs, collies, etc. Following elabo-ration theory's method of sequencing concepthierarchies,dog, being lower in the hierarchy,is considered a more detailed concept thanmammal; hence, dog should not be taughtuntil the concept of mammal has been pre-sented. This prescription, of course, fails totake into account children's existing knowl-edge structure that includes dog at A youngage but not mammal. A more naturalsequencing method would start with dog andproceed to the unfamiliar technical conceptsof mammals, vertebrates, etc.

Another problem with current models istheir relative neglect of the job of integratingdifferent content elements in learners' minds.As described above, current models breakdown content into different types of learningoutcomes, and prescribe different learningconditions for each type. This analyticapproach builds in a bias toward teachingdisconnected, isolated content elements out oftheir naturally occurring contexts. Becauserules or skills are often taught separately fromfacts or verbal information, their interrelated-ness tends to be de-emphasized. This is unfor-tunate, because rules are best taught inmeaning-rich environments, and verbalinformation is best taught in meaningfulcontexts that include action and problem-solving performance. In fact, different kindsof outcomes seem to depend on each other foroptimal learning environments. Teaching onekind of content in exclusion of other kindscreates problems for both learner and teacher.Elaboration theory addresses the problem byrecommending periodic synthesizers andsummarizers, but its sequencing strategycenters around one kind of learning outcome(concept, procedure, or principle) for an entire

course. Thinking of content structure only interms of concept hierarchies, procedures, orcausal models is overly restrictive in light ofcognitive research suggesting the importanceof other kinds of knowledge structure (Strike &Posner, in press).

Some of the teaching models being otferedby cognitive researchers bear strong resem-blance to traditional ID models. Larkin andChabay (1989), for example, offer designguidelines for the teaching of science in theschools (pp. 160-163):

1. Develop a detailed description of theprocesses the learner needs to acquire.

2. Systematically address all knowledgeincluded in the description of process.

3. Let most instruction occur throughactive work on tasks.

4. Give feedback on specific tasks assoon as possible after an error is made.

5. Once is not enough. Let studentsencounter each knowledge unitseveral times.

6. Limit demands on students' attention.By any standard, these design guidelines

are very c/ose to the prescriptions found incomponent display the( ry, elaboration theory,and Gagne's instructional-design theory. Thestrong correspondence can be scen as goodnews for instructional-design theories: Currentcognitive researchers seem to agree on somefundamentals of design that also form thebackbone of ID models.

On the other hand, some teaching modelsrecently developed and tested emphasizedesign elements that traditional ID modelshistorically have under-emphasized. Thecognitive apprenticeship model includes somewell-worn design elements Auch as modelingand fadingald others that a,e relativelyneglected by ID modelssuch as situatedlearning, exploration, and the role of tacitknowledge.

FEATURES OF

COGNITIVE APPRENTICESHIPS

In the section below, the design elementsof the cognitive apprenticeship model (basedprimarily on Ccllins, 1991) are reviewed andrelated to traditional ID concepts.

1. Content: Teach tacit, heuristic knowl-edge as well as textbook knowledgeCollins et al. (1989) refer to four kinds ofknowledge that differ somewhat from IDtaxonomies:


Domain knowledge is the conceptual,factual, and procedural knowledge typi-cally found in textbooks and otherinstructional materials. This knowledgeis important, but often is insufficient toenable students to approach and solveproblems independently.Heuristic strategies are "tricks of thetrade" or "rules of thumb" that oftenhelp narrow solution paths. Expertsusually pick up heuristic knowledgeindirectly through repeated problem-solving practice; however, heuristicknowledge can be made explicit andtaught directly.Control strategies are required forstudents to monitor and regulate theirprobiem-solving activity. Controlstrategies have monitoring, diagnostic,and remedial components; this kind ofknowledge is often termed metacogni-lion in the literature (Paris & Winograd,1990).Learning strategies are strategies forlearning ; they may be domain, heuris-tic, or control strategies, aimed at learn-ing. Inquiry teaching to some extentdirectly models expert learning strate-gies (Collins & Stevens, 1983).

CommentID taxonomies (Gagne, 1985; Merrill, 1983)

pertain primarily to domain or textbook knowl-edge, although the distinction is not explicit.Gagne's "cognitive strategies" fits the lastthree strategies listed by Collins et al.Merrill's "find a principle" or "find a proce-dure" seems closest to those three strategies.Both Gagne and Merrill are least specificabout instruction of this type, although theyboth acknowledge its importance.

2. Situated learning: Teach knowledgeand skills in contexts that reflect theway tne knowledge will be useful inreal life. 5rown, Collins, and Duguid(1989) argue for placing ail instructionwithin "authentic" contexts that mirrorreal-life problem-solving situations. Collins(1991) is less forceful, moving away fromreal-life requirements and toward problem-solving situations: for teaching math skills,situated learning could encompass settings"ranging frcm running a bank or shoppingin a grocery store to inventing new theo-rems or finding new proofs. That is, situ-ated learning can incorporate situations

from everyday life to the most theoreticalendeavors" (Collins, 1991, p. 122).

Collins differentiates a situated-learn-ing approach from common schoolapproaches:

We teach knowledge in an abstract wayin schools now, which leads to strategies,such as depending on the fact thateverything in a particular chapter usesa single method, or storing informationjust long enough to retrieve it for a test.Instead of trying to teach abstractknowledge and how to apply it incontexts (as word prob!ems do), weadvocate teaching in multiple contextsand then trying to generate across thosecontexts. In this way, knowledgebecomes both specific and general.(Collins, 1991, p. 123)Collins cites several benefits for placing

instruction within problem-solvingcontexts:

Learners learn to apply their knowledgeunder appropriate conditions.Problem-solving situations foster inven-tion and creativity (see also Perkins,1990).Learners come to see the implicationsof new knowledge. A most commonproblem inherent in classroom learningis the question of relevance "How doesthis relate to my life and goals?" Whenknowledge is acquired in the context of5olving a meaningful problem, thequestion of relevance is at least partlyanswered.Knowledge is stored in ways that makeit accessible when solving problems.People tend to retrieve knowledge moreeasily when they return to the setting ofits r.,cquisition. Knowledge learnedwEile solving problems gets encoded ina way that can be accessed again inproblem-solving situations.

Although not cited by Collins andcolleagues, two other cognitive teachingmodels are relevant to the notion of situatedlearning. Bransford and colleagues(Bransford, Sherwood, Hasselbring, Kinzer, &Williams, 1990) have developed an approachcalled "anchored instruction." They take aSherlock Holmes or Indiana Jones videodiscand develop problem-solving activities aroundincidents in the video. The instruction is thusgrounded in a rich mw:ro-context that ismeaningful and interesting to the learner.They have pointed out many similarities tosituated learning and cognitive apprentice-


ships (The Cognition and Technology Group atVanderbilt, 1990). The second related model isSpiro's cognitive flexibihty theory (Spiro,Coulson, Feltovich, & Anderson, 1988; Spiro &Jehng, 1990) which grew cut of studies ofmedical students learning advanced subjectmatter. Spiro and colleagues found that forstudents to avoid oversimplifying complexcontent, they needed to see multiple analogiesacross multiple contexts. Spiro and colleaguesused a series of "mini-cases" to help studentsa mental model that was sensitive to the manynuances and subtleties of the content. BothI3ransford's anchored instruction and Spiro'scognitive flexibility theory stress the impor-tance of placing problem-solving instructionwithin meaningful contexts.

CommentTaken in its extreme form that would

require "authentic," real-life contexts for alllearning, the notion of situated learning issomewhat vague and unrealistic. Instructionalways involves the dual goals of generaliza-tion and differentiation (Gagne & Driscoll,1989). In its more modest form, however, theidea of context-based learning has consider-able appeal. Gagne & Merrill (1990) havepointed to the need for better integration oflearning goals through problem-solving"transactions." The notion of situated learn-ing, however it is viewed, challenges theconditions-of-learning paradigm thatprescribes the breaking down of tasks to betaught out of context.

3. Modeling and explaining: Show how aprocess unfolds and tell reasons why ithappens that way. Collins (1991) cites twokinds of modeling: (1) modeling ofprocesses observed in the world and (2)modeling expert performance, includingcovert cognitive processes. Computers canbe used to aid in the modeling of theseprocesses. Collins stresses the importanceof integrating both the demonstration andthe explanation during instruction.Learners need access to explanations asthey observe details of the modeledperiormance. Computers are particularlygood at modeling covert processes thatotherwise would be difficult to observe,both natural and mental. Collin5 suggeststhat truly modeling expert performance,including the false starts, dead ends, andbackup strategies, can help learners morequickly adopt the tacit forms of knowledgealluded to above under the Content

section. Teachers in this way are seen as"intelligent novices" (Bransford, Coin,Hasselbring, Kinzer, Sherwood, drWilliams, 1988). By seeing both processmodeling and accompanying explanations,students can develop "conditionalized"knowledge, that is, kr lwledge about whenand where knowledge should be used tosol-fe a variety of problems.

CommentID models presently incorporate modeling

and demonstration techniques. Tying expla-nations to modeled performances is a usefulidea, similar to Chi and Bassok's (1989) studiesof worked-out examples. Again, the emphasisis on making tacit strategies mere explicit bydirectly modeling mental heuristics.

4. Coaching: Observe students as theytry to complete tasks and provide hintsand helps when needed. Intelligenttutoring systems sometimes embodysophisticated coaching systems that modelthe the learner's progress and providehints and support as practice activitiesincrease in difficulty. Burton and Brown(1982) developed a coach to help learnerswith "How the West Was Won," a gameoriginally implemented on the PLATOsystem. Anderson, Boyle, and Reiser (1985)developed coaches for geometry and LISPprogramming.

Coaching strategies can be imple-mented at least partially in traditionalschool settings. Bransford and Vye (1989)identify several characteristic of effectivecoaches:

Coaches need to monitor learners'performance to prevent their gettingtoo far off base, but leaving enoughroom to allow for a real sense of explo-ration and problem solving.Coaches help learners reflect on theirperformance and compare it to others'.Coaches use problem-solving exercisesto assess learners' knowledge states.Misconceptions and buggy strategiescan be identified in the context of solv-ing problems; this is particularly true ofcomputer-based learning environments(Larkin & Chabay, 1989).Coaches use problem-solving exercisesto create the "teaching moment."Posner, Strike, Hewson, and Gertzog(1982) present a 4-stage model forconceptual change: (1) studentsbecome dissatisfied with their miscon-


ceptions; (2) they come to a basicunderstanding of an alternative view; (3)the alternative view must appear plausi-ble; and (4) they see the new view'svalue in a variety of new situations (seealso Strike Sr Posner, in press).

CornmentCoaching probably involves the most

"instructional work" (cf. Bunderson dr Inouye,1987) of any of the cognitive apprenticeshipmethods. Short of one-on-one tutoring, coach-ing is likely to be partial and incomplete.Cooperative learning and small-group learn-ing methods can provide some coachingsupport for individual performance. Andcomputers can help tremendously in monitor-ing learner performance and providirg real-time helps; yet presently coaching is only fullyimplemented in resource-intensive ir. elligenttutoring systems. Much work is being done tomodel the essentials of coaching functions oncominiter systems; we continue to needresource-efficient methods for achieving thecoaching function,

5. Articulation: Have students thinkabout their actions and give reasonsfor their decisions and strategies, thusmaking their tacit knowledge more explicit.Think-aloud protocols are one example ofarticulation (Hayes St Flower, 1980; Smith,1988). Collins (1991) cites the benefits ofadded insight and the ability to compareknowledge across contexts. As learners'tacit knowledge is brought to light, thatknowledge can be recruited to solve otherproblems.

CommentJohn Anderson (1990) h3s shown that

procedural knowledgethe kind that peoplecan gain automaticity inis initially encodedas declarative or conceptual knowledge, butlater fades as the skill br-omes procedural-ized. Methods for articulating tacit knowledgehelps to restore a conscious awareness ofthose lost strategies, enabling more flexibleperformance. Traditional ID models suggestpracticing problem solving to learn problemsolving, but are surprisingly lacking in specificmethods to teach learners to think consciouslyabout covert strategies.

6. Reflection: Have students look backover their efforts to complete a taskand analyze their own performance.Reflection is like articulation, except it is

pointed backwards to past tasks. Analyzingpast performance efforts can also involveelements of strategic goal-setting andintentional learning (Bereitcr &Scardamalia, 1989). Collins and Brown(1988) suggest four kinds or levels ofreflection:

Imitation occurs when a batthig coachdemonstrates a proper swing, contrast-ing it with your swing;Replay occurs when the coach video-tapes your swing and plays it back,critiquing and comparing it to the swingof an expert;Abstracted replay might occur bytracing an expert's movement of keybody parts such as elbows, wrists, hips,and knees, and comparing thosemovement to your movements;Spatial reification would take thetracings of body parts and ph.. themmoving through space.

The latter forms of reflection clearly rely ontechnologiesvideo or computer forinstantiation. Collins (1991) uses Andersonet al.'s Geometry Tutor as an example nfreflective instrtiction.

CommentArticulation and reflection are both strate-

gies to help bring meaning to activities thatmight otherwise be more "rote" and procedu-ral. Reigeluth's (1983) concern with meaning-ful learning is indicative of the need; however,much of traditional ID practice tends to da-value the reflective aspects of performance infavor of getting the procedure down right.

7. Exploration: Encourage students to tryout different strategies and hypothe-ses and observe their effects. Collins(1991) claims that through exploration,students learn how to set achievable goalsand to manage the pursuit of those goals.They learn to set and try out hypotheses,and to independently seek knowledge.Real-world exploration is always an attrac-tive option; however, constraints some-times prohibit extensive time in realisticsettings. Simulations are one way to allowexploration; hypertext structures areanother.

CommentAs Reigeluth (1983) notes, discovery learn-

ing techniques are less efficient than directinstruction techniques. The choice mustdepend, to some extent, on the goals of


instruction: for near transfer tasks (cf. Salomon& Perkins, 1988), direct instruction may occa-sionally be warranted; for far transfer tasks,learners must learn not only the content butalso now to solve unforeseen problems usingthe content; in such cases, instructionalstrategies allowing exploration and strategicbehavior become essential.

Having thus represented a traditional IDmode of thinking about exploration, I am stiPleft unsatisfied. Following a cognitive-appren-ticeship way of thinking, there is somethingintrinok :y valuable about situated problem-se lying activity that makes learning workbetter than straight procedural practice. Thisis an area that needs better articulationamong ID theorists.

8. Sequence: .Present instruction in anordering of simple to complex,increasing diversity, and global beforelocal skills.

Increasing complexity. Collins et al.(1989) point to two methods for helpinglearners deal with increasing complex-ity. First, instruction should take stepsto control the complexity of assigt.edtasks. They cite Lave's study of tii,loringapprenticeships: apprentices first Warnto sew drawers, which have straightlines, few pieces of material, and nospecial features like zippers or pockets.They progress to more complexgarments over a period of time. Thesecond method for controlling complex-ity is through scaffolding; for example,group or teacher support for individualproblem solving.Increasing diversity refers to the varietyin examples and practice contexts.Global before local skills refers tohelping learners acquire a mentalmodel of the prublem space at veryearly stages of learning. Even thoughlearners are not engaged in full prob-lem solving, through modeling andhelping on parts of the task(scaffolding), they can understand thegoals of the activity and the way variousstrategies relate to the problem's solu-tion. Once they have a clear"conceptual map" of the activity, thcycan proceed to developing greater skillat specific skills.

CommentThe sequencing suggestions above bear a

strong resemblance to those of elaborationtheory (Reigeluth & Stein, 1983) and compo-nent display theory (Merrill, 1983). The notionof global before local skills is implicit in elabo-ration theory; simple-to-comp.x sequencingis the foundation of elaboratior. theory. Thenotion of increasing diversity is the near-equivalent to the prescription to use "variedexample" and practice activities in concept orrule learning. The cognitive apprenticeshipextends these notions beyond rule learning toproblem-solving contexts.

EXAMPLES OF

COGNITIVE APPRENTICESHIP

In the section below, I briefly review threeapproaches that Collins et al. (1989) identify asembodying cognitive apprenticeship features.

Procedural Facilitations for WritingNovice writers typically employ a

knowledge-telling strategy: they think abouttheir topic, then write their thought down;think again, then write the next thought down,and so on until they have exhausted theirthought: about the topic. This strategy, ofcourse, is in conflict with a more constructive,planning approach in which writing pieces arccomposed in a more coherent, intentional way.To encourage students to adopt more sophis-ticated writing strategies, Scardamalia andBereiter (1985) have developed a set of writingprompts called Procedural Facilitations, thatare designed to reduce working-memorydemands and provide a structure for complet-ing writing plans and revisions. Their systemincludes a set of cue cards for differentpurposes of writing, structured under fiveheadings: new idea, improve, elaborate, goals,and prat' it together. Each prompt is writtenon a notecard and drawn by learners workingin small groups. The teacher makes use of twotechniques, soloing and co-investigation.Soloing gives learners the opportunity to tryout new procedures by themselves, thenreturn to the group for critique and sugges-tions. Co-investigation is a process of usingthink-aloud protocols that allow learner andteacher to work together on writing activities.This allows for more direct modeling andimmediate direction. Bereiter andScardamalia (1987) have found up to tenfoldgains in learning indicators with nearly every


learner improving his/her writing through theintervention.

Reciprocal TeachingBrown te Palinscar (1989) have developed a

cooperative learning system for the teachingof reading, termed Reciprocal Teaching. 71,,..!

teacher and learners assemble in groups of 2to 7 and read a paragraph together silently. Aperson assumes the "teachee role and formu-lates a question on the paragraph. Thisquestion is addressed by the group, whosemembers are playing roles of producer andcritic simultaneously. The "teachee advancesa summary, and makes a prediction or clarifi-cation, if any is needed. The role of teacherthen rotates, and the group proceeds to thenext paragraph in the text. Brown andcolleagues have also developed a method ofassessment, called dynamic assessment,based on successively increasing prompts.The Reciprocal Teaching method uses acombination of modeling, coaching, scaffold-ing, and fading to achieve impressive results,with learners showing dramatic gains incomprehension, retention, and far transferover sustained periods.

Schoenfeld's Math TeachingSchoenfeld (1985) studied methods for

teaching math to college students. He devel-oped a set of heuristics tnat were hdpful insolving math problems. His method intro-duces those heuristics, as well as a set ofcontrol strategies and a productive beliefsystem about math, to students. Like thewriting and reading systems, Schoenfeld'ssystems includes explicit modeling of prob-lem-solving strategies, and a series of struc-tured exercises affording learnIr practice inlarge and small groups, ;".s well as individually.He employs a tactic he calls "postmortemanalysis," retracing the solution of recentproblems, abstracting out the generalizablestrategies and component!. Unlike the writingand reading systems, Schoenfeld carefullyselects and sequences practice cases to movelearners into higher levels of skill. Anotherinteresting technique is the equivalent to"stump the teacher," with time at the begin-ning of each class period devoted to learner-generated problems that the teacher ischallenged to solve. Learners witnessingoccasional false starts and dead ends of theteacher's solution can acquire a more appro-priate belief structure about the nature ofexpert math problem solving. Schoenfeld'spositive research findings support a growing

body of math research suggesting the impor-tance o5 acquiring a conceptual or schema-based representation of math problemsolving.

Forecasting ApprenticeshipsCOMET (Cooperative Program for

Operational Methodology Education andTraining) is an inter-agency program thatdevelops weather training for forecasters affil-iated with the National Weather Service, theAir Weather Command, and the Navy.Interactive videodisc training modulesdeveloped for these forecasters are also madeavailable to university meteorology depart-ments across the nation. COMET's distancelearning program is presently nearing betatesting of its first two modules, DopplerInterpretation and Convection Initiation. TheCollins-Brown cognitive apprenticeship modelprovides the conceptual foundation for themodule series. Each module uses opticallaserdisc, CD-ROM, and object-orientedprogramming to simulate a forecaster work-station and provide forecasting practice underrepresentative conditions. The elements ofthe cognitive apprenticeship model arediscussed below as they relate to the design ofthe Convection Initiation module.

Content. Cognitive apprenticeships teachdomain or textbook knowledge, but they alsomake explicit the strategic knowledge neededto use that knowledge in solving real prob-lems. A key goal for the COMET module wasto provide authentic activities that requiredthe forecaster to use situation-specific knowl-edge to make a forecast. There are no distinct"concept" lessons or "rule" lessons. Instead, aseries of forecasting problems are presentedwith accompanying "conceptual models" andprocedural "tutorials." In short, the domainand strategic content is made available to thelearner in the immediate context of solvingspecific forecasting problems.

Methods. Computer-based instruction(CBI) is relatively well-suited to cognitiveapprenticeship teaching methods. In additionto the modeling and scaffolding common toCBI, a growing number of toolkit-styleprograms and simulations allow for degrees ofexploration. Reflectioncomparing onesown problem-solving processes with others'is possible though somewhat t'ifficult with CBI.The COMET modules replicate a sophisti-cated forecasting workstation, where bothlearner and program may initiate actionstoward problem solution, presentation, orfeedback. However, because the module


lacks a natural language interface, the inter-action is somewhat constrained. A continuingchallenge for CBI design is to develop meth-ods of making instruction approach a continu-ous dialogue rather than a forced-choice, pre-fabricated path.

Sequence. As noted above, the sequenc-ing rules of 3 cognitive apprenticeship maydepend on the type of content being taught--e.g., hypertext exploration for schema building(Wilson & Jonassen, 1989) versus simple-to-complex case selection for skill building (cf.Reigeluth & Stein, 1983). The ConvectionInitiation module utilizes a progression offorecasting cases in a simple-to-complexorder, similar to Schoenfeld's approach tomath teaching; at the same time, a variety ofhypertext structuresdata, rules, definitions,etc.are available to the learner on demandusing hypertext access structures. This is to beexpected: often a mix of well-structuredproblems and hypertext-like exploration isprobably optimal for problem-solvinginstruction.

Sociology. The social aspects of thecognitive apprenticeship model seem to posethe greatest challenge for CBI design. Fromwhat we know about learning, social variablesare powerful mediators in learning (e.g.,Eckert, 1990; Salomon, Globe: son, &Guterman, 1989). CBI is often administeredon an individual basis, with a single learnerengaged at a computer workstation. Althoughgroup work is a desirable option, sometimes itis not possible. The COMET weather forecast-ing modules are a case in point. Modules areto be completed on the job, but because fore-casting offices are sometimes stalled with only2 - 3 persons on duty at a time, learners mustcomplete instruction individually during breaktimes while others continue on the job. Thedesign challenge is: how can we build inneeded social reinforcement and convey asense of community in what is essentially anindividualized learning experience? Toaddress the cognitive apprenticeship's socialdesign, the Convection Initiation moduleincorporated several strategies, including:

1. The rrodule begins with a forecastingoffice scenario. Th -.! learner is teamedwith "Ron," a fellow forecaster, andimmediately given a forecastingproblem to solve. This same officescenario is used as a wrap-up, where thelearner completes a 2-hour forecastingshift containing four forecasts.

2. Following the beginning .4f:cc scenario,the learner is briefly introduced to a

panel of three forecasting experts.These experts serve as guides throughthe module.

3. Following each forecasting practice, oneof the three experts provides feedbackon the case in the form of an "expertanswer." This is a 30 - 90 second expla-nation of the case. Meteorological dataon-screen are highlighted with arrows;this is combined with either a CD audiooverlay or audio-video sequence of theexpert using a chromaboard, much likea TV weatherperson explaining a map.The module's beta testing will becompletzd over the spring and sum-mer; we will report different effects ofaudio-only vs. chromaboard explana-tions. In either case, the personalizedfeedback from the actual expertsshould allow some sodal modeling tooccur. The meteorological field is fairlysmall; moreover, there are no annualconventions typically attended bymembers of the field. Getting to knowthe forecasting experts through the IVDmodules should help strengthen thesense of community within the field.

4. As stated above, the module concludeswith an office-based sImulation wherethe forecaster, teamed with "Ron,"solves a series of realistic, time-basedforecasting problems.

Through these efforts to make the modulemore personalized, the social elementsneeded to support good instruction are artifi-cially created. We believe these compensat-ing strategies will hep improve the receptionof the modules in the field, and improve theattitudes of learners completing the moduleson an individual basis.

SUMMARY AND CONCLUSION

This paper has been concerned with therelationship between two related disciplines:cognitive psychology and instructional design.ID, the more applied discipHne, faces thechallenges of constantly re-inventing itself asnew research in basic psychology sheds lightson learning processes. It is no cause for con-cern that ID models need continual self-review; indeed, there would be cause forconcern if the situation were otherwise. Toprovide some sense of context, however, Iwould like to call attention to some basic prin-ciples of cognitive apprenticeship articulatedby the philosopher Erasmus more than 500


years ago. In a letter to a student friend,Erasmus offers some advice:

Your first endeavor should be to choose themost learned teacher you can find, for it isimpossible that one who is himself noscholar should make a scholar of anyoneelse. As soon os you find him, make everyeffort to see that he acquires the feelings ofa father towards you, and you in turn thoseof a ,on towards him Second:y, youshould give him attention and be regular inyour work for him, for the talents of stu-dents are sometimes ruined by violenteffort, whereas regularity in work haslasting effect just because of its temper-ance and produces by daily practice agreater result than you would :,uppose .... Aconstant element of enjoyment must bemingled with our studies so that we think oflearning as a game rather than a form ofdrudgery, for no activity can be continuedfor long if it does not to some extent affordpleasure to the participant.Listen to your teacher's explanations notonly attentively but eagerly. Do not besatisfied simply to follow his discourse withan alert mind; try now and then to antici-pate the direction of his thought .... Writedown his most important utterances, forwriting is the most faithful custodian ofwords. On the other hand, avoid trusting ittoo much .... The contests of minds in whatwe may call their wrestling ring are espe-cially effective for exhibiting, stimulating,and enlarging the sinews of the humanunderstanding. Do not be ashamed to askquestions if you are in doubt; or to be putright whenever you are wrong.Erasmus (1974, pp. 114-115)At a time when terms like metacognition,

scaffolding , and cognitive apprenticeship arebeing invented to describe the learningprocess, it is humbling to be reminded of theinsights of former paradigms. Instructional-design theorists may chafe at the continuingneed to revise their theories in light ofadvances in psychological theory, but it isgood for both fields for the dialogue tocontinue. Indeed, the interaction betweenbasic rsychology and applied instructionaldesign can be expected to continue for sometime to come.

1 1

REFERENCES

Anderson, J. R. (1990). Cognitive psychologyand its implications (3rd ed.). New York:Freeman.

Bereiter, C., & Scardamalia, M. (1987). Thepsychology of written composition.Hillsdale, NJ: Erlbaum.

Bereiter, C., & Scardamalia, M. (1989).Intentional learning as a goal of instruction.In L. B. Resnick (Ed.), Knowing, learning,and instruction: Essays in honor ofRobert Glaser. Hillsdale, NJ: Erlbaum.

Bransford, J. D., Coin, L. I., Hasselbring, T. S.,Kinzer, C. K., Sherwood, R. D., & Williams, S.(1988). Learning with technology:Theoretical and empirical perspectives.Peabody Journal of Education,64 (1), 5-26.

Btansford, J. D., Sherwood, R. D., Hasselbring,T. S., Kinzer, C.K., & Williams, S. M. (1990).Anchored instruction: Why we need it andhow technology can help. In D. Nix & R.Spiro (Eds.), Cognition, education, andmultimedia: Exploring ideas in highteesnology. Hillsdale, NJ: Erlbaum.

Bransford, J. D., & Vye, N. J. (1989). A perspec-tive on cognitive research and its implica-tions for instruction. In L. B. Resnick & L. E.Klopfer (Eds.), Toward the thinkingcurriculum: Current cognitiveresearch. 1989 ASCD Yearbook.

Bunderson, C. V., & Inouye, D. K. (1987). Theevaluation of computer-aided educationaldelivery systems. In R. M. Gagne (Ed.),Instructional technology: Foundations.Hillsdale, NJ: Erlbaum.

Collins, A. (1991). Cognitive apprenticeshipand instructional technology. In L. Idol & B.F. Jones (Eds.), Educational values andcognitive instruction: Implications forreform. Hillsdale, NJ: Erlbaum.

Collins, A., & Brown, J. S. (1988). The computeras a tool for learning through reflection. InH. Mandl & A. Lesgold (Eds.), Learningissues for intelligent tutoring systems.New York: Springer.

Collins, A., Brown, J. S., & Newman, S. E. (1989).Cognitive apprenticeship: Teaching thecrafts of reading, writing, and arithmetic. InL. B. Resnick (Ed.), Knowing, learning,and instruction: Essays in honor ofRobert Glaser. Hillsdale, NJ: Erlbaum.

Collins, A., & Stevens, G. (1983). Inquiryteaching. In C. M. Reigeluth (Ed.),Instructional-design theories andmodels: An overview of their currentstatus. Hillsdale, NJ: Erlbaum.


Eckert, P. (1990). Adolescent socialcategoriesInformation and sciencelearning. In J. Gardner, J. G. Greeno, F. Reif,A. H. Schoenfeld, & A. A. diSessa, & E. Stage(Eds.), Toward a scientific practice ofscience education. Hillsdale, NJ: Erlbaum,

Erasmus, D. (1974). The correspondence ofErasmus. Volume 1: Letters 1 to 141,1484 to1500. Translated by R. A. B. Mynors and D.F. S. Thomson. Toronto: University ofToronto Press.

Gagne, R. M. (1985). The condiiions oflearning (4th ed.). New York: Holt,Rinehart, & Winston.

Gagne, R. M., dc Driscoll, M. P. (1989).Essentials of learning for instruction(2nd ed.). Englewood Cliffs, NJ: Prentice-Hall.

Gagne, R. M., & Merrill, M. D. (1990).Integrative goals for instructional design.Educational Technology Research andDevelopment,38 (1), 23-30.

Glaser, R. (1982) Instructional psychology:Past, present, end future. AmericanPsychologist, 37 (2), 292-305.

Glaser, R., & Bassok, M. (1989). Learningtheory and the study of instruction. AnnualReview of Psychology,40, 631-666.

Gott, S. P. (1988, April). The lostapprenticeship: A challenge forcognitive science. Paper presented at themeeting of the American EducationalResearch Association, New Orleans.

Hayes, J. R., & Flower, L. S. (1980). Identifyingthe organization of writing processes. In L.W. Gregg & E. R. Sternberg (Eds.),Cognitive processes in writing.Hillsdale, NJ: Lawrence ErlbaumAssociates.

Merrill, M. D., Kowallis, T., & Wilson, B. G.(1981). Instructional design In transition.In F. Farky & N. Gordon (Eds.),Psychology and education: Thestate of the union. Chicago:McCutcheon.

Palinscar, A. M., & Brown. A. L. (1984).Reciprocal teaching of comprehension-fostering and monitoring activities.Cognition and Instruction, 1, 117 - 175.

Paris, S. C., & Winograd, P. (1990). Howmetacognition can promote academiclearning and instruction. In B. F. Jones & L.Idol (Eds.), Dimensions of thinking andcognitive instruction. Hillsdale, NJ:Erlbaum.

Perkins, D. N. (1990). The nature and nurtureof creativity. In In B. F. Jones & L. Idol (Eds.),Dimensions of thinking and cognitiveinstruction. Hillsdale, NJ: Erlbaum.

Posner, G. K., Strike, K. A., Hewson, P. W., SECertzog, W. A. (1982) Accommodation ofscientific conception: Toward a theory ofconceptual change. Science Education, 66,211-227.

Reigeluth, C. M. (Ed.). (1983). Instructional-design theories and models: Anoverview of their current status.Hillsdale, NJ: Erlbaum.

Reigeluth, C. M. Meaningfulness andinstruction: Relating what is being learnedto what a student 1. ows. -nstructi analScience, 12(3) 197-218.

Reigeluth, C. M., & Stein, R. (1983). Elaborationtheory. In C. M. Reigeluth (Ed.),Instructional-design theories andmodels: An overview of their ctarentstatus. Hillsdale, NJ: Erlbaum.

Reigeluth, C. M., Merrill, M. D., Wilson, B. G.,& Spiller, R. T. (1980). The elaborationtheory of instruction: A model forstructuring instruction. InstmctionalScience, 9, 195 - 219.

Reiser, R. A. (1987). Instructional technology: Ahistory. In R. M. Gagne (Ed.).,Instructional techno:ogy: Foundations.Hillsdale, NJ: Erlbaum.

Resnick, L. B. (1981). Instructional Psychology.Annual Review of Psychology.

Resnick, L. B. (1989, February). Learning inschool and out, Educational Researcher, 5- 8.

Rogoff, B. (1990). Apprenticeship inthinking: Cognitive development insocial context. New York: OxfordUniversity Press.

Rogoff, B., & Lave, J. (Eds.). (1984). Everydaycognition: Its development in socialcontext (pp. 95 - 116). Cambridge, MA:Harvard University Press.

Salomon, G., Globerson, T., & Guterman, E.(1989). The computer as a zone of proximaldevelopment: Internalizing reading-relatedmetacognitions from a reading partner.Journal of Educational Psychology, 81(4),620 -627.

Salomon, G., Sr Perkins, D. N. (1989). Rockyroads to transfer: Rethinking mechanisms ofa neglected phenomenon. EducationalPsychologist, 24(2), 113-142.

Scardamalia, M., & Bereiter, C. (1985).Fostering the development of self-regulation in children's Knowledgeprocessing. In S. F. Chipman, J. W. Segal, &R. Glaser (Eds.), Thinking and learningskills: Research and opei; questions (pp.563 - 577). Hillsdale, NJ: Erlbaum.


Schoenfeld (1985). Mathematical problemsolving. New York: Academic Press.

Smith, P. (198 ). Think-aloud protocol paper.Spiro, R. J., Coulson, R. L., Feltovich, P. J., &

Anderson, D. K. (1988, October).Cognitive flexibility theory: Advancedknowledge acquisition in ill-structureddomains (Technical Report No. 441).Champaign, IL: University of Illinois atUrbana-Champaign, Center for the Studyof Reading.

Spiro, R. J., & Jehng, J-C. (1990). Cognitiveflexibility and hypertext: Theory andtechnology for the nonlinear andmultidimensional traversal of complexsubject matter. In D. Nix & R. J. Spiro (Eds.),Cognition, education, and multimedia:Exploring ideas in high technology (pp.163-205). Hillsdale, NJ: Lawrence ErlbaumAssociates.

Strike, K. A., & Posner, G. J. (in press). Arevisionist theory of conceptual change. Toappear in R. Duschl & R. Hamilton (Eds.),Philosophy of science, cognitivescience, and educational theory andpractice. Albany, NY: SUNY Press.

Th2 cognition and technology group atVanderbilt. (1990, August-September).Anchored instruction and its relation tosituated cognition. EducationalResearcher, 2-10.

Wertsch, J. V. (1989). Vygotsky and thesocial formation of mind. Cambridge,MA: Harvard University Press.

Wilson, B. G., & Jonassen, D. H. (1989).Hypertext and instructional design: Somepreliminary guidelines. PerformanceImprovement Quarterly, 2(3), 34 - 49.

Wood, D. J. (1980). Teaching the young child:Some relationships between socialinteraction, language, and thought. In D. R.Olson (Ed.), The social foundations oflanguage and thought. New York: Norton.

document resume ed 335 022 ir 015 189 author wilson, … · 2013-08-02 · document resume ed 335...

Documents