8/16/2019 Retrieval-Based Learning; Active Retrieval Promotes Meaningful LearningXXX

1/12

Downloaded from cdp.sagepub.com at GEORGE MASON UNIV on October ! "#$%

Retrieval-Based Learning: ActiveRetrieval Promotes MeaningfulLearning

Je rey D. Kar ic!ePurdue University

A"stractRetrieval is the key process for understanding learning and for promoting learning, yet retrieval isnot often granted the central role it deserves. Learning is typically identi ed with the encoding orconstruction of knowledge, and retrieval is considered merely the assessment of learning thatoccurred in a prior experience. The retrieval !ased learning perspective outlined here is groundedin the fact that all expressions of knowledge involve retrieval and depend on the retrieval cuesavaila!le in a given context. "urther, every time a person retrieves knowledge, that knowledge ischanged, !ecause retrieving knowledge improves one#s a!ility to retrieve it again in the future.Practicing retrieval does not merely produce rote, transient learning$ it produces meaningful, longterm learning. %et retrieval practice is a tool many students lack metacognitive awareness of anddo not use as often as they should. &ctive retrieval is an e'ective !ut undervalued strategy forpromoting meaningful learning.

If you know something, or if you have stored informa-tion about an event from the distant past, and never usethat information, never think of it, your brain is func-

tionally equivalent to that of an otherwise identical brain that does not “contain” that information.

— ndel !ulving "#$$#%

!o understand learning, it is essential to understand the pro-cesses involved in retrieving and reconstructing knowledge.&e may think we know something, that our minds contain or

pos- sess some knowledge, but the only way to assessknowledge is by engaging in an act of retrieval. 'ifferencesin the ability to recover knowledge may not stem from whatis “stored” in our minds but rather from differences in theretrieval cues available in particular conte(ts. )iven the

fundamental importance of retrieval for understanding the process of learning, it is surpris- ing that retrieval processeshave not received more attention in educational research.*onsider that over the past decade, many influential +ational

esearch *ouncil books on how people learn havecontained no mention of retrieval "+ational esearch *ouncil,

, /a, /b%.It is essential to consider retrieval processes not only

because they are central to understanding learning butalso because the act of retrieval itself is a powerful tool for enhanc- ing learning. 0oreover, active retrieval does notmerely produce rote, transient learning1 it produces

meaningful, long- term learning. !he idea that retrieval isthe centerpiece for

http://cdp.sagepub.com/http://cdp.sagepub.com/http://cdp.sagepub.com/http://cdp.sagepub.com/

8/16/2019 Retrieval-Based Learning; Active Retrieval Promotes Meaningful LearningXXX

2/12

Downloaded from cdp.sagepub.com at GEORGE MASON UNIV on October ! "#$%

understanding learning, coupled with the importance of active retrieval for producing learning, is referred to asretrieval- based learning .

Learning Based on t#e Design of t#eMind&e often think of our minds as places in our heads,mental spaces or containers where we store knowledge.

oediger "#$2 % noted that for centuries, most metaphorsused to describe mental processes have characteri3ed themind as a physical space and knowledge as physicalthings in that space—for e(ample, by likening our minds

to libraries filled with books or cabinets loaded with files"see too 0oscovitch, 4%. In education, the metaphor of a physical building is often used to describe the mind andknowledge. 5nowledge is constructed by learners whoactively build knowledge struc- tures 1 researchers seek tounderstand the architecture of the mind1 and instructors aidstudents by providing scaffolding for learning.

&hen minds are viewed as places for storing knowledge,it is natural to focus attention on processes involved inconstructing new knowledge in storage. ducational research

http://cdp.sagepub.com/http://cdp.sagepub.com/http://cdp.sagepub.com/http://cdp.sagepub.com/

8/16/2019 Retrieval-Based Learning; Active Retrieval Promotes Meaningful LearningXXX

3/12

() Karpic

and instructional practices have placed a premium onidentify- ing the best ways to encode knowledge ande(periences. etrieval processes, the processes involved inusing available cues to actively reconstruct knowledge,have received less attention. !here seems to be a tacitassumption that successful encoding or construction of

knowledge, in itself, is sufficient to ensure learning.6asic research on learning and memory, however, hasemphasi3ed that retrieval must be considered in any analysisof learning. In part, this is because people do not store static,e(act copies of e(periences that are reproduced verbatim atretrieval. Instead, knowledge is actively reconstructed on the

basis of the present conte(t and available retrieval cues"6artlett, #$7 1 +eisser, #$841 see too 0oscovitch, 41

oediger, %. !he reconstructive nature of mind isrevealed in the systematic errors people make in retrievingknowledge, errors that verbatim recording devices wouldnot make. !he past never occurs again in its e(act form, soa mental store- house of copies of past e(periences would

be of little use. 9eople instead have the ability to use the past to meet the demands of the present by reconstructingknowledge rather than reproducing it e(actly.

&hat people e(press when they reconstruct knowledgedepends on the retrieval cues available in a given conte(t.:lti- mately, knowledge reconstruction depends on thediagnostic value of cues, the degree to which cues help

people recover particular target information to thee(clusion of competing candidates "+airne, 1

aai;makers < =hiffrin, #$2#%. &e may wish to e(aminewhat a person has constructed or stored in mind, but it isimpossible to directly assess the contents of storage, per se.&e can only ever e(amine what a person reconstructsgiven the available cues and conte(t " oediger, 1

oediger < )uynn, #$$81 !ulving < 9earlstone, #$88%.!hus, it is essential to consider retrieval processes in anyanal- ysis of learning.

!he second crucial reason retrieval is important for learn-ing is that learning is altered by the act of retrieval itself.

very time a person retrieves knowledge, that knowledge ischanged, because retrieving knowledge improves one>sability to retrieve it again in the future "5arpicke <

oediger, 4, 21 5arpicke < ?aromb, # %. !his is afeature of a func- tional learning and memory system. @ur minds are sensitive to the likelihood that we will need

knowledge at a future time, and if we retrieve something inthe present, there is a good chance we will need to recover it again. !he process of retrieval itself alters knowledge inanticipation of demands we may encounter in the future.

etrieval is therefore not only a tool for assessing learning but also a tool for enhancing learning " oediger <5arpicke, 8a%.

Re eated Retrieval$n#ances Long-%ermLearning

Imagine you are studying for an upcoming e(am. After youhave read through your notes or your te(tbook one time, what

8/16/2019 Retrieval-Based Learning; Active Retrieval Promotes Meaningful LearningXXX

4/12

() Karpicwould you want to do ne(tB Cou have three optionsD Cou can

"a% go back and restudy either all of the material or parts of it,"b% try to recall the material without restudying afterward, or "c% do something else. &hich would you chooseB

&e "5arpicke, 6utler, < oediger, $% gave thisques- tion to a large group of college students. 0oststudents "/4E% said they would reread their notes or

te(tbook, and #E said they would do something else.@nly #2E said they would attempt to recall materialafter reading it. # !he decision to repeatedly read makessense if we identify learning with pro- cesses of encodingand constructing knowledge and consider retrieval to beonly a way to assess prior learning. It stands to reason thatmore studying "i.e., more encoding and knowledgeconstruction% should produce more learning, whereasretrieval should measure learning but not produce it.

&ould students be better off repeatedly reading thanengag- ing in retrievalB &e conducted an e(periment witha design that mirrored the question asked in the survey" oediger < 5arpicke, 8b%. =tudents read educationalte(ts and recalled them under one of three conditions.@ne group of students spent time repeatedly studying ate(t in four study periods. A second group read a te(t inthree study periods and then recalled it in one retrieval

period "labeled === %, in which the students wrote down asmany ideas from the te(t as they could recall. A third groupread the te(t during one study period and then practicedrecalling it during three consecutive repeated retrieval

periods. =tudents did not reread the te(t or receive any

feedback after any of the recall periods1 they only prac-ticed actively retrieving material.

At the end of the learning phase, the students made a judg- ment of learning D a prediction of how well they wouldremem- ber the material in the future. !hen, one week later,students recalled the material again to see how much theyactually retained in the long term.

Figure #b shows students> ;udgments of learning. !hemore times students repeatedly read the material, the

better they believed they had learned it. Gowever, Figure #ashows that students> actual learning e(hibited the opposite

pattern. !he more times students practiced actively retrievingthe material, the better they retained it in the long term.=tudents spent the same amount of time e(periencing thematerial in all three conditions, and students in therepeated-retrieval condition only recalled and did notrestudy the te(t, yet active retrieval produced the best long-term retention "for further discussion of metacognitiveawareness of the effects of retrieval practice, see 5arpicke <)rimaldi, # %.

eturning to the survey of student learning strategies"5arpicke et al., $%, one might think the results wouldchange if we reworded the survey question. +amely, studentsmight choose to engage in active retrieval if they couldreread after attempting retrieval. In a second version of thesurvey, we asked students the e(act question describedabove, but changed option "b% to say, “try to recall thematerial, and then go back and restudy the te(t.” =tudents>choices did change, but not as dramatically as one mighte(pectD Forty-two percent

8/16/2019 Retrieval-Based Learning; Active Retrieval Promotes Meaningful LearningXXX

5/12

Retrieval-Based ()

a Actual Recall

.&

.'

.%

.

.(

b )red*cted Recall%.#

.+

.'

.

."

.#

(.+

."SSSS SSSR SRRR (.'

SSSS SSSR SRRR

&ig. '. "inal recall a- after repeatedly studying a text in four study periods condition- , reading a text in three study periods and then recalling it in one retrievalperiod R condition-, or reading a text in one study period and then repeatedlyrecalling it in three retrieval periods RRR condition-. /udgments of learning !- weremade on a 0 point scale, where 0 indicated that students !elieved they wouldremem!er material very well. The data presented in these graphs are adapted from1xperiment 2 of Roediger and 3arpicke 2445!- . The pattern of students#metacognitive 6udgments of learning predicted recall- was exactly the opposite of the pattern of students# actual long term retention.

of students said they would practice retrieval and then reread.Gowever, H#E of students still said they would only reread,and #4E said they would do something else. In other words,/2E of students indicated that they would not practice activeretrieval even when they would have the opportunity toreread afterward.

!he benefits of active retrieval are large when studentsretrieve and then reread. In part, this is because attemptingretrieval improves students> encoding when they restudymate- rial, a phenomenon known as the “potentiating” effectof retrieval "see )rimaldi < 5arpicke, in press1 5arpicke,

$1 5ornell, Gays, < 6;ork, $1 &issman, awson,< 9yc, ##%. In an e(periment that demonstrated the

power of retrieving plus rereading, we "5arpicke <

oediger, # % had students practice retrieval of educational te(ts about sci- entific topics. !here wereseveral conditions in this e(peri-

.,

.+

.&

.'

.%

.

.(

."

.$

.#Stud- Retr*e e

Once

RepeatedRetr*e al

ment, but three particular conditions are relevant to thisdiscussion. In one condition, students read a te(t once in asingle study period, whereas students in a second conditionread the te(t, recalled as much as they could in a recall

period, and then reread the te(t briefly. In a third condition,students repeatedly recalled the te(t across a series of eightalternating study recall periods "four study periods and four recall periods in total%. @ne week later, the studentsrecalled the material again to assess long-term retention.

Figure shows the proportion of ideas recalled one week after original learning session. 9racticing retrieval one time doubled lonterm retention relative to reading the te(t

) r o p o r t * o

n o

f I d e a s

R e c a

l l e d

/ u d g m e n t o

f 0 e a r n

* n g

) r o p o r t

* o n o

f I d e a s

R e c a

l l e d

8/16/2019 Retrieval-Based Learning; Active Retrieval Promotes Meaningful LearningXXX

6/12

Retrieval-Based ()&ig. (. Long term retention after studyingonce, practicing retrieval once followed !yrereading-, or practicing repeated retrieval.

The data presented in this graph are adaptedfrom 1xperiment 2 of 3arpicke and Roediger

24(4-. Practicing retrieval one time dou!ledlong term retention, and repeated retrievalproduced a 7448 improvement in retentionrelative to studying once.

once "7HE vs. #/E%, and engaging in repeated retrievalincreased retention to 2 E. !hus, practicing active retrievalwith brief rereading between recall attempts, in whatamounted to about a 7 -minute learning session, producedlarge benefits for long-term learning "see too 0c'aniel,Goward, < instein, $%.

8/16/2019 Retrieval-Based Learning; Active Retrieval Promotes Meaningful LearningXXX

7/12

(5 Karpic

a Verbat*m 1uest*ons b Inference 1uest*ons

.+

.&

.'

.%

.

.(

."

.$

.#Stud- Repeated 2onceptRetr*e al

.+

.&

.'

.%

.

.(

."

.$

.#Stud- Repeated 2oncept Retr*e al

Stud-Mapp*ng )ract*ce Stud- Mapp*ng )ract*ce

c Metacogn*t* e )red*ct*ons.+

.&

.'

.%

.

.(

."

.$

.#Stud- Repeated 2oncept Retr*e al

Stud- Mapp*ng )ract*ce

&ig. ). Proportion correct on nal short answer ver!atim 9uestions a- and inference 9uestions !-( week after learning, and metacognitive 6udgments of learning predicted proportion of itemscorrect- made during the initial learning phase c-. Practicing retrieval enhanced long term learningrelative to ela!orative studying with concept mapping, yet this !ene t was largely unanticipated !ystudents. Reprinted from :Retrieval Practice Produces ;ore Learning Than 1la!orative tudying !y /. ?. 3arpicke and /. R. @lunt, 24((, Science, 331 , pp. 002A00). =opyright 24((

!y the &merican &ssociation for the &dvancement of cience. Reprinted with permission.

Active RetrievalPromotes MeaningfulLearning

Acurrent challenge is to establish the effectiveness of retrieval- based learning activities with educational materialsand assess- ments that reflect comple(, meaningful learning.“0eaningful learning” is often defined in contrast to “rotelearning” "0ayer, 2%. &hereas rote learning is brittle andtransient, meaning- ful learning is robust and enduring. ote

learning is thought to produce poorly organi3ed knowledge that laccoherence and integration, which is reflected in failures to mainferences and transfer knowledge to new problems. 0eaningful learning, in contrast, is thought to produce organi3ed, coherent, and

) r o p o r t

* o n 2 o r r e c

t

) r o p o r t

* o n

2 o r r e c

t

/ u d g m e n

t o

f 0 e a r n

* n g

8/16/2019 Retrieval-Based Learning; Active Retrieval Promotes Meaningful LearningXXX

8/12

(5 Karpicintegrated mental models that allow people tomake inferences and apply knowledge.

It is important to remember that, in allcircumstances, peo- ple transfer past e(periencesto meet the demands of a unique present. !hisalways involves reconstructing knowledge byusing the cues available in a given retrievalconte(t. @utcomes identified as “rote” or

“meaningful” learning may not reflect differences in whatlearners have encoded, stored, or con- structed. Instead,the distinction between rote and meaningful learning hingesupon the similarity between retrieval scenarios in the presentand learning e(periences from the past. !he abil- ity to useknowledge in the present depends on the diagnostic value of retrieval cues regardless of whether the goal of

8/16/2019 Retrieval-Based Learning; Active Retrieval Promotes Meaningful LearningXXX

9/12

Retrieval-Based (5

a Actual )erformance b Metacogn*t* e )red*ct*ons

.% .%

. .

.( .(

."2oncept

Mapp*ng

Retr*e al)ract*ce

."2onceptMapp*ng

Retr*e al)ract*ce

&ig. *. Proportion correct on a nal concept map test a- and metacognitive 6udgments of learning predicted proportion of items correct- during the initiallearning phase !-. The data presented in these graphs are adapted from1xperiment 2 of 3arpicke and @lunt 24((!-. 1ven though the criterial testinvolved creating a concept map, practicing retrieval during original learningenhanced performance relative to ela!orative studying with concept mapping,despite the fact that students !elieved ela!orative concept mapping wouldproduce !etter learning.

retrieval is to recall a fact, make an inference, or solve a new problem.

Investigators have taken two general approaches to e(am-ine the effects of active retrieval on meaningful assessments.@ne approach has been to use final-assessment questions that

differ from questions e(perienced during original learning"e.g., 6utler, # 1 *han, $1 *han, 0c'ermott, <oediger, 81 Gin3e < &iley, ##1 Johnson < 0ayer,

$1 ohrer, !aylor, < =holar, # %. A second approachhas been to design assessments that measure students>abilities to make infer- ences, apply knowledge, and solvenew problems, or that oth- erwise measure the coherenceand integration of students> mental models. &e havealready seen that active retrieval enhances learning of meaningful educational materials and that these effects arelong-lasting, not short-lived. ecent research has shown that

practicing active retrieval enhances performance asmeasured on meaningful assessments of learn-

ing and that these effects can be greater than those produced by other active-learning strategies.

In a recent study, we "5arpicke < 6lunt, ##b% e(aminedthe effects of active retrieval using measures of meaningfullearning1 importantly, we compared these effects to those pro-duced by a popular learning strategy known as concept mapping "+ovak < )owin, #$2H%. *oncept mappinginvolves having students create diagrams in which conceptsare represented as nodes and relations among concepts arerepresented as links connecting the nodes. !his activityinvolves elaborative study- ing because it is thought to helpstudents organi3e and encode meaningful relationshipsamong concepts. *oncept mapping can also be used as a tool

to assess students> knowledge, as dis- cussed in the follow paragraphs. *oncept mapping is

) r o p o r t

* o n

2 o r r e c

t

/ u d g

m e n

t o

f 0 e a r n

* n g

8/16/2019 Retrieval-Based Learning; Active Retrieval Promotes Meaningful LearningXXX

10/12

Retrieval-Based (5a popular strategy and en;oys strong advocacyamong many educators. Indeed, the activitywould seem to be an effective tool for

promoting elaborative processing, although therehas not been a wealth of randomi3ed, controllede(periments e(am- ining the most effective waysto use concept mapping as a learn- ing activity"see 5arpicke < 6lunt, ##a%.

&e had students read science te(ts andcreate a concept map or practice activelyretrieving the ideas from the te(ts. In two controlconditions, students read the material once or repeatedly. @ne week after the learning phase,the students answered two types of questionsdesigned to assess meaning- ful conceptuallearningD verbatim questions , which assessedconceptual knowledge directly included in thete(t, and infer- ence questions , which requiredstudents to make connections across concepts.

As shown in Figures 7a and 7b, practicing retrieval produced the best performance on both types of con- ceptualquestions, even better than elaborative studying withconcept mapping did. Gowever, when students made predic-tions of their long-term learning during the initial learning

phase "Fig. 7c%, they believed that rereading and conceptmap- ping would produce more learning than active retrievalwould, even though the opposite was true.

In a second e(periment, we again had students either create concept maps or practice active retrieval while theystudied. !his time, however, as our criterial assessment of long-term learning, we had students create a concept map,which itself is a method of assessing the coherence andintegration of stu- dents> conceptual knowledge. ven onthis final concept-map assessment "Fig. Ha%, students

performed best when they had engaged in active retrievalduring learning, a result indicating that active retrievalenhanced students> deep, conceptual

8/16/2019 Retrieval-Based Learning; Active Retrieval Promotes Meaningful LearningXXX

11/12

(5 Karpic

learning of the material. Cet once again, the studentsgenerally believed they would do better after elaborativestudying than after practicing retrieval, as shown in FigureHb. 0ost students did not e(pect active retrieval to producemore learning than elaborative studying with conceptmapping, but in fact it did.

Retrieval-Based Learning: Re riseetrieval is the key process for understanding learning and

for promoting learning. It is essential for understandinglearning because all e(pressions of knowledge involveretrieval and therefore depend on the retrieval cues that areavailable in a given conte(t. !he diagnostic value of retrieval cues—the degree to which cues help peoplerecover particular target knowledge to the e(clusion of competing candidates—is the critical factor for all learning.

etrieval is the key to promoting learning, and activeretrieval has powerful effects on long- term learning. ach

act of retrieval alters the diagnostic value of retrieval cuesand improves one>s ability to retrieve knowl- edge again inthe future. etrieval may enhance learning because itimproves the match between a cue and particular desiredknowledge, or it may enhance learning by constraining thesi3e of the search set—the set of potentially recoverablecandidates that comes to mind in the conte(t of a cue"5arpicke < 6lunt, ##b1 5arpicke < ?aromb, # %. 9rac-ticing retrieval has also been shown to enhance organi3ational

processing "*ongleton < a;aram, in press1 ?aromb <oedi- ger, # %, and such processing is likely necessary to

support performance on assessments of meaningfullearning. !hus, there are a number of potential mechanisms

by which active retrieval may enhance long-term learning.etrieval-based learning is a broad, general perspective on

how to improve student performance. !here are manylearning activities that active retrieval could potentially beincorporated into, and there are many different ways in whichretrieval could be integrated into such activities. For instance, group discus- sions, reciprocal teaching, andquestioning techniques "both formal ones, such as providingclassroom qui33es, and infor- mal ones, such as integratingquestions within lectures% are all likely to engage retrieval

processes to a certain e(tent. =pend- ing time activelyattempting to retrieve and reconstruct one>s knowledge is a

simple yet powerful way to enhance long-term, meaningfullearning. !he central challenge for future research will be tocontinue identifying the most effective ways to use retrievalas a tool to enhance meaningful learning.

Recommended Reading

5arpicke, J. '., < )rimaldi, 9. J. " # %. "=ee eferences%. A recentoverview of retrieval-based learning that covers additional top-ics, including metacognitive aspects of retrieval practice and the

potentiating effects of retrieval on subsequent encoding.0oscovitch, 0. " 4%. "=ee eferences%. An accessible critique of

the storehouse metaphor of memory that also emphasi3es theessential role of retrieval in understanding learning.

oediger, G. K. " %. "=ee eferences%. An in-depth discussion of why retrieval is the key process for understanding learning.

oediger, G. K., < 5arpicke, J. '. " 8a%. "=ee eferences%. Acom- prehensive and historical overview of research on theeffects of active retrieval on learning.

!ulving, . "#$$#%. "=ee eferences%. ssential reading for any

student of learning and memoryD an interview in which !ulv-ing e(plains why retrieval processes must be considered in anyanalysis of learning.

Ac!no+ledgments

!he author would like to thank =tephanie 5arpicke for assistancewith preparing the manuscript.

Declaration of ,on icting nterests!he author declared that he had no conflicts of interest with respectto his authorship or the publication of this article.

&unding

!he writing of this paper was supported by +ational =cienceFoundation )rant ': - $H##4 and by the :.=. 'epartment of

ducation>s Institute of ducation =ciences )rant 7 /A## $ 7.!he opinions e(pressed here are those of the author and do not rep-resent views of the Institute of ducation =ciences or the :.=.'epartment of ducation.

/ote#. &e were unable to score ambiguous responses given by HE of

students.. !here is some evidence that retrieving a portion of some material

can impair recovery of nonpracticed parts of the material "=torm,##%, but with meaningful and integrated educational materials,

practicing retrieval of part of the materials typically enhances learn-ing of nonpracticed parts of the materials "*han, $1 *han,0c'ermott, < oediger, 8%.

References6artlett, F. *. "#$7 %. Remembering: A study in experimental

and social psychology . *ambridge, nglandD *ambridge:niversity 9ress.

6utler, A. *. " # %. epeated testing produces superior transfer of learning relative to repeated studying. Journal of

Experimental Psychology: earning! "emory! and #ognition!$% , ###2L##77.

*han, J. *. 5. " $%. &hen does retrieval induce forgetting andwhen does it induce facilitationB Implications for retrievalinhibi- tion, testing effect, and te(t processing. Journal of

"emory and anguage! %& , #/7L#4 .*han, J. *. 5., 0c'ermott, 5. 6., < oediger, G. K. " 8%.

etrieval-induced facilitationD Initially non-tested material can benefit from prior testing of related material. Journal of Experi- mental Psychology: 'eneral! &$( , //7L/4#.

*ongleton, A., < a;aram, =. "in press%. !he origin of theinteraction between learning method and delay in the testingeffectD !he roles of processing and conceptual retrievalorgani3ation. "emory ) #ognition .

8/16/2019 Retrieval-Based Learning; Active Retrieval Promotes Meaningful LearningXXX

12/12

Retrieval-Based (5

)rimaldi, 9. J., < 5arpicke, J. '. "in press%. &hen and why doretrieval attempts enhance subsequent encodingB "emory )#ognition .

Gin3e, =. ., < &iley, J. " ##%. !esting the limits of testing effectsusing completion tests. "emory! &* , $ L7 H.

Johnson, *. I., < 0ayer, . . " $%. A testing effect with multi-

media learning. Journal of Educational Psychology! &+& , 8 #L 8 $.5arpicke, J. '. " $%. 0etacognitive control and strategy selectionD

'eciding to practice retrieval during learning. Journal of Experi- mental Psychology: 'eneral! &$, , H8$LH28.

5arpicke, J. '., < 6lunt, J. . " ##a%. esponse to comment on“ etrieval practice produces more learning than elaborativestudying with concept mapping.” cience! $$. , H/7.

5arpicke, J. '., < 6lunt, J. . " ##b%. etrieval practice producesmore learning than elaborative studying with concept mapping.

cience! $$&, 44 L44/.5arpicke, J. '., 6utler, A. *., < oediger, G. K. " $%. 0etacogni-

tive strategies in student learningD 'o students practice retrieval

when they study on their ownB "emory! &/ , H4#LH4$.5arpicke, J. '., < )rimaldi, 9. J. " # %. Retrieval-based learning:

A perspective for enhancing meaningful learning . :npublishedmanuscript, 9urdue :niversity, &est Kafayette, I+.

5arpicke, J. '., < oediger, G. K. " 4%. epeated retrieval duringlearning is the key to long-term retention. Journal of "emoryand anguage! (/ , #/#L#8 .

5arpicke, J. '., < oediger, G. K. " 2%. !he critical importanceof retrieval for learning. cience! $&*, $88L$82.

5arpicke, J. '., < oediger, G. K. " # %. Is e(panding retrieval asuperior method for learning te(t materialsB "emory )#ogni- tion! $, , ##8L# H.

5arpicke, J. '., < ?aromb, F. 0. " # %. etrieval mode distin-guishes the testing effect from the generation effect. Journal of "emory and anguage! %0 , 4L 7$.

5ornell, +., Gays, 0. J., < 6;ork, . A. " $%. :nsuccessfulretrieval attempts enhance subsequent learning. Journal of

Experimental Psychology: earning! "emory! and #ognition!$( , $2$L$$2.

0ayer, . . " 2%. earning and instruction " nd ed.%. :pper =ad-dle iver, +JD 9earson.

0c'aniel, 0. A., Goward, '. *., < instein, ). @. " $%. !heread- recite-review study strategyD ffective and portable.

Psychologi- cal cience! 0+ , /#8L/ .0oscovitch, 0. " 4%. 0emoryD &hy the engram is elusive. In

G. K. oediger, C. 'udai, < =. 0. Fit3patrick " ds.%, cience of memory: #oncepts "pp. #4L #%. +ew Cork, +CD @(ford :niver-sity 9ress.

+airne, J. =. " %. !he myth of the encoding-retrieval match. "em- ory! &+ , 72$L7$/.

+ational esearch *ouncil. " %. 1o2 people learn: 3rain!mind! experience! and school . J. '. 6ransford, A. K. 6rown, <

. . *ocking " ds.%. &ashington, '*D +ational Academies9ress.

+ational esearch *ouncil. " /a%. 1o2 students learn: ciencein the classroom . 0. =. 'onovan < J. '. 6ransford " ds.%.

&ash- ington, '*D +ational Academies 9ress. +ational esearch *ouncil. " /b%. 4a5ing science to school:

earning and teaching science in grades 6-, . . A. 'uschl,G. A. =chweingruber, < A. &. =house " ds.%. &ashington, '*D

+ational Academies 9ress. +eisser, :. "#$84%. #ognitive psychology . +ew Cork, +CD Appleton-

*entury-*rofts. +ovak, J. '., < )owin, '. 6. "#$2H%. earning ho2 to learn . *am-

bridge, nglandD *ambridge :niversity 9ress.aai;makers, J. ). &., < =hiffrin, . 0. "#$2#%. =earch of

associative memory. Psychological Revie2! ,, , $7L#7H.oediger, G. K. "#$2 %. 0emory metaphors in cognitive psychology.

"emory ) #ognition! , , 7#L H8.oediger, G. K. " %. &hy retrieval is the key process in under-

standing human memory. In . !ulving " d.%, "emory! con- sciousness! and the brain: 4he 4allinn conference "pp. / L4/%.9hiladelphia, 9AD 9sychology 9ress.

oediger, G. K., < )uynn, 0. J. "#$$8%. etrieval processes. In .K. 6;ork < . A. 6;ork " ds.%, "emory "pp. #$4L 78%. =an'iego, *AD Academic 9ress.

oediger, G. K., < 5arpicke, J. '. " 8a%. !he power of testingmemoryD 6asic research and implications for educational prac-tice. Perspectives on Psychological cience! & , #2#L # .

oediger, G. K., < 5arpicke, J. '. " 8b%. !est enhanced learningD!aking memory tests improves long-term retention. Psychologi-

cal cience! &/ , H$L //.ohrer, '., !aylor, 5., < =holar, 6. " # %. !ests enhance the trans-

fer of learning. Journal of Experimental Psychology: earning! "emory! and #ognition! $% , 77L 7$.

=torm, 6. *. " ##%. !he benefit of forgetting in thinking and remem- bering. #urrent 7irections in Psychological cience! 0+ , $#L $/.

!ulving, . "#$$#%. Interview. Journal of #ognitive 8euroscience! $ ,2$L$H.

!ulving, ., < 9earlstone, ?. "#$88%. Availability vs. accessibility of information in memory for words. Journal of 9erbal earning and 9erbal 3ehavior! ( , 72#L7$#.

&issman, 5. !., awson, 5. A., < 9yc, 0. A. " ##%. !he interimtest effectD !esting prior material can facilitate the learning of new material. Psychonomic 3ulletin ) Revie2! &, , ##H L##H4.

?aromb, F. 0., < oediger, G. K. " # %. !he testing effect in freerecall is associated with enhanced organi3ational processes.

"emory ) #ognition! $, , $$/L# 2