ED 064 692

ANTHCRTITLF

SPJNS AGr,NCYBUREAU NOPUB DATFCONTRACTNOTr,

EDRS PR1CFDCSCRIPTCR;;

ABSIPACT

DOCUMENT RESUME

24 CS 000 067

Levine Joel R.Scme Thoughts About Cognitive Strategies and RoalinqComprehension. Theoretical Paper No. 30.Wisconsin Univ., Madison. Research and DevelopprCentc.r tor Cognitive Learning.Office of Education (DBEW), Washington, D.C.PP-5-02I6Dec 7/OEC-5-10-15415p.

MF-$0.65 fiC-$3.29Cognitive Processes; *Individual Differences;*Psychological Studies; Reariing Ability; *Reaj!in.;Comprehonsion; *Reading Research

Psychological experiwents investigating imposed andinduced cognitive strategis are reviewed And related to operatisin reading comprehension. it has been suggestei that comprehensiondifferences between good and poor readers, may arise from the way inwhich they haLitually organize intra- and inter-sentence elementsduring input. Subject-generated visual imagery is singled out asparticularly effective organizational _.-trategv. Irplications of thisresearch are considered in the context of aptitude by treatmentinteractions and individual differences. (Author)

Thuoretical l'apQr

SOW 7HOUGHTS Al3C1 T COGNITIVE STRATEGIES

Verbal

HerbertThomas A

CCNITPEHENSICN

by Joel P., Levin

S OF. (PARTMENT OF HU ALTHkCJUCATiO14 & Wti_F AREOF FtCE OF (EDUCA1FON

::)( kIMENT 0-44S Eitf PIPA)Dut ID I xAt.:11. AS RICEIVI-.0 FROM

PF c4SON OR ORIJAhb2A:ION URIC!NA, 17 POINT Of VII 1.1, 74 OP: NJ

,iTAIID Do NOT l'4ICP::-S4R(t.RF PRF SIN T OcrtrIA: Cif k IC F WI,

A T.I f05.11:.)P4 OR Pat 0.:1

Repo.-t from the Project on Variables andProcesses ]n Cognitive Learning

and Visual Components of Children's Learning

J. Klausrneier, Robert E. Davidson, Joel R. Levin,. Romberg, B. Robert Tabac.Lnick, Alan M. Voelker,

Larry Wilder, Peter WolffProect Investigators

Mary R. QuillingQuality Verification Program Director

Dorothy A. FrayerAssistant Sc.-i,-ntist

Wisconsin Research and DevelopmentCenter for Cognitive LearningThe University of Wisconsin

Madison, Wisconsin

December 1971

Published by the Research and :development Centel for 0.1.x:five 1.cainine Sltpportedin part tis 8 research and development center by hinds from the United States Office of Edueation,Department of Health, rducation, and Welfare. The opinions expressed herein de not neciessardvreflect :he position or policy of the Office of Eductition and no official endorsemeat b !ht. Officeof Education should be inferred.

Center No. C-03 / Contract OE 5-10-154

NATIONAL EVALUATION COMMITTEESamuel Browns!!P ofessor of Urban tducationGrachiate SchoolVale University

Launor F. CarterSenior Vice President an

Technology and DevelopmentSystem Development Corporation

Francis S. ChaseProfessorDeportment of EducationUniversity of Chicago

Henry ChaunceyPres (fentEducational Testing Service

Martin DeutschDirector. Institute far

Developmental StudiesNew Yark Mother,' College

.k.ck EdlingDirector, Teaching Research

DivisionOreaon State System of Higher

Education

Elizaboth KoontzWale and Labor Standards

Administrotion. U.S.Derailment of labor,Wushington

Roderick McPheePresidentPunahou School, Honolulu

G. Wesley SowardsDirector. Elementary EducationFlorida State University

Patrick Suppe%ProfessorDepartment of MathematicsStanford University

*Benton J. UnderwoodProfessorDepartment of PsychologyNorthwestern University

RESEARCH AND DEVELOPMENT CENTER POLICY REVIEW BOARD

Leonard BerkowitzChairmanDeportment of Psychology

Archie A. BuchmillerDeputy Stat. SuperintendentDeportment of Public Instruction

Robert E. GrinderChairmanDeportment of Educational

Psychology

Russell J. HosierProfeuer, Curriculum

and Instruction

Clauston JenkinsAsvstant DirectorCoordinating Committee for

Higher Education

Herbert J. KlausmeiorDirector, R & D CenterProfessor of Educational

Psychology

EXECUTIVE COMMITTEE

Edgar F. BorgattoBrittinghari Professor of

Sociology

Anne E. BuchananProwl SpecialistR & D (enter

Robin S. ChapmanResearch AssocicnoR & 0 Center

Stephen C. KleineDean, College of

Letters and Science

Donald J. McCartyDeanSchool of Education

Ira SharkanskyAssociate Professor of PoliNcal

Science

B. Robert TabachnickChoirmon, Deportment

of Cuetkulunl andInstruction

Henry C. WeinIkkExecutive SecretaryWisconsin Education Association

M. Crawford YoungAssociate DeanThe Graduate Schaal

Robert E. DavidsonAssistant Professor,

Educational Psychology

Frank H. FarloyAssociate Professor,

Educat!onal Psychology

Russell J. HosierProfessor of Curnculum and

Instruction ond of Business

*Herbert J. KlausmelorDirector, R & D CenterProfessor of Educational

Psychology

Wayne OttoProfessor of Curriculum and

InstructrOn Ph/cycling)

Robort G. PetzoldAssociate Dean of the School

of EducotionProfessor of Curriculum and

Instruction ond of Music

FACULTY OF PRINCIPAL INVESTIGATORS

Vernon L. AllenProfessor of Psychology

Ted CzaikowskiAssistant Professor of Curriculum

ond Ins"ructlain

Robert E. DavidsonAssistant Professor of

Educational Psychoiiny

Frank H. FarleyAtsociate ri..4emor of Educational

Psychology

Lester S. Golubtscrunsr in Curricvlum and

instruction and in Engiish

John G. HarveyAssociate Professor of

Mothernatics and of rorriculumand Instruction

James MoserAssistant Professor of Mathematics

Education; Visiting Scholar

Wayne OttoProfessor of Curricufum ond

Instruction IReading)

Milton 0. PellaProfaner of Curriculum and

instruction IScientell

Richard L. VenezkyAssistant Professor of Engfish

and of Computer Sciences

Alan VottIkerAssistant Professor of Curriculum

and Instruction

Lorry WilderAssistant Professor of Curriculum

and Instruction

Gary A. Davis Herbert J. Klausmeier Thomas A. Romberg Peter WolffAssociate fvofessor of Director, R & 0 Cantor Associate Director, R & D Center Assistant Professor of Educational

fduiorionni Psychologi Professor of Educational Professor of Mothernotscs and of PsydiologyPsychology Curriculum and Instruction

M. Vere DeVaultProfessor cif Curriculum ono

instruction IMatnematia0

Donald LangeAssatont PrOfelSOr of Curt culom

and Instruction

MANAGEMENT COUNCIL

Herbert J. KlausmelerDirector. ft & C. Center

.A.C. Hammon Professor ofEducational Psychology

Mary R. QuillingDirector

Technical Development Program

B. Robert TabachnkkCha.rman, Department

of CurAculurn andinstruction

Thomas A. RombergAssociate Director

II

James WalterDiroctor

Dissemination Program

Dan G. WoolportDirector

Omrations and Business

COMMITTII OIASSINAN

Statement of Focus

The Wisconsin Research and Development Center for Cognitive Learningfocuses on contributing to a better understanding of cognitive learning by chil-dren and youth and to the improvement of related educational practices. Thestrategy for research and development is comprehensive. It ircludes basic re-search to generate new knowledge about the conditions and processes of learn-ing and about the processes of instruction, and the subsequent development ofresearch-based instructional materials, many of which are designed for use byteachers and others for use by students. These materials are tested and refinedin school settings. Throughout these operations behavioral scientists, curricu-lum experts, academic scholars, and school people interact, insuring that theresults of Center activities are based soundly on knowledge of subject matterand cognitive learning and that they are applied to the improvement of educa-tional practice.This Working Paper is irom the Project on Variables and Processes in Cog-nitive Learning in Program I, Conditions and Processes of Learning. Generalobjectives of the Program are to generate knowledge about concept learning andcognitive skills, to synthesize existing knowledge and develop getseral taxon-omies. models, or theories of cognitive learning, and to utilize the knowledgein the development of curriculum materials and procedures. Contributing tothese Program objectives, this project has these objectives: to ascertain theimportant variables in cognitive learning and to apply relevant knowledge tothe development of instructional taaterials and to the basic processes and abil-ities involved in e -..ncept learning; and to develop a system of individually

guided motivation for use in the elementary nhool.

iii

iv

Acknowledgments

I am grateful to Mrs. Linda Monahan and Miss Terry Fiscus for typing the

final draft of the paper.

I

Contents

Page

List of rigures vi

Abstract vii

Introduction

imposed Characteristics and Reading Comprehension

Induced Characteiistics and Reading Comprehension 4

References '7

List of Figures

Figuie Page

1 Hypothesized ordinal interaction between aptitude (population)and treatment (imposed independent variabl .) to account forsome recent tindings

vi

A bstract

Psychological experimentL, investigating imposed and induced cognitivestrategies are reviewed and related to oper,ations in reading comprehension.It has been suggested that comprehension differences between good and poorreaders may arise from the way in which they habitually organize intra- andinter-sentence elements during input. Subject-generated visual imagery issingled out as a particularly effective organizational strategy. Implicationsof this research are considered in the context of aptitude by treatment inter-actions and individual differences.

vil

Introduction

I do not profess to be an expert in read-ing. Although this sentence is ambiguous, itapplies to both my personal reading speedand comprehension, as well as my familiaritywith the reading and perception literature inthe respective fields of education and psy-chology. What I would like to do, as an out-sider looking in, is to share with you somepotential educational applications that grewout of a symposium entitled "Issues in imageryand learning" held at the 1971 Western Psy-chological Association meeting in San Fran-cisco. The primary focus of the symposiumwas on visual imagery in children, and itsreported positive relationship with learningand memory.

The bulk of the experimentation in whichimagery processes have been engaged andinferred (through the use of pictorial andimage-evoking materials) has incorporatedassociative learning tasks in laboratory set-tings. In a few experiments, comprehensionof sentences and sentence-embedded materialshas been investigated. A recumng result isthat materials which are concrete, imageable,and dynamic are easier to remember than thosewhich are not (e.g., Paivio. 1969; Rohwe:-,1967),

A sentence is assumed to make its con-stituents more memorable by virtue of theorganization it bestows on them. It has teenhypothesized that the success of the organi-zation depends upon the extent to which in-creased contextual meaning and imagery isproduced (Levin, 1971). A collection of f:en-tences should impose an organization on itsconstituents in analogous fashion. In thispaper, we will restrict our attention to rea-sonably concrete materials. That is, sen-tences like "Continuous fraud negates impliedsincerity" are probably less concrete andimageable than sentences like "Giddy spin-sters terrify squealing infants" (Davidson,19(.6). Paivio (1970) has presented datawhich support this notion, by showing that

the fozmer type of sentence is accompanioiby lrnger imagery latencies (i.e., a greateramount of time is required to form a mentalimage of the sentence's contents), as well asby inferior recall of the general meaning ofthe sentence (though not necessarily the in-dividual words).

Consider the school-age child who can-not comprehendand consequently will notrememberthe content of what he reads. I

am not referring to the child who cannot iden-tify (decode) the words, although the ensuingdiscussion may in fact be relevant when con-sidering this type of reading disability aswell.' Neither am I referring to the child whocan identify the words correctly, but cannotderive meaning from them because they areforeign to his experiential vocabulary (seeWiener & Cromer. 1967). For now, I will focusmy attention on the child who can identify thewords, knows the meaning of individual words,but has difficulty in integeating the separatemeanings into an organized whole. The childto whom I am referring possesses average orabove.average decoding and vocabulary skills,but exhibits poor performance on tasks whichinvelve reading comprehension.

Wiener and Cromer (1967) have consideredthis type of reading disability in what theyhave called a "difference" model. Unlike thetraditional view that all reading problems re-sult frpm either disorders (generally organic)or deficits (lack of prerequisite identificationand/or vocabulary skills), these authors haveargued that at least two other models of read-ing difficulty need to be considered: the "dis-ruption' model, where emotional and psycho-logical barriers which are interfering with thereading process must be removed; and the"difference" model.

'Personal communication from Dr. RogerA. Severson, Associate Professor of Educa-tional Psychology, University of Wisconsin,February 1971.

1

'Inc "difference" mudel aSsorts that:

...reading difficulty is attributable to'it renr cs or mismatches between the

mode of responding and thatwhich is more appropriate, and thus hasthe best payoff in a particular situation.This model assumes that the individualwould read adequately if the materialwere consistent with his behavior pat-terns; thus, a chan,,:e in either the ma-eria1 or in his patterns of verbalization

is a prerequisite for better reading.[V\ iener & Cron:er, I . 62-91

ake another look at the last sentence..% hat this is :-aying is that a "difference"between good and poor readers is attributableto the way in which they respectively input

hat they read. Good readers typically corn--ehend; poor readers do not. In order forpoor readers to perform more like good readers,one of two events must occur: (a) the readingmaterials must be changed in some way (e.g.,their content, their structure, their representa-tional mode, and the like); or (b) poor readersmust learn (be taught) to employ some of thesuccessful responding "habits of goodreaders. It is interesting that the same tworecommendations have been made elsewhere,with regard to making the performance ofchildren who are poor learners more like thatof children who are good learners (e.g., I evin,Rohwer, & Cleary, 1971; Rohwer, 1970).

At the imagery and learning symposium,I recommended that a distinction be madebetween two different approaches to the studyof facilitative variables in children's learning(Levin, 1971). One line of research has typi-cally dealt with manipulations of learningmaterials which render them more or lessmemorable. Specifically, verbal and imaginalrepresentations imposed on subjects by ex-peri!...enters have been shown to affect per-formance on learning tasks. Rohwer's (1967)

extensive investigation of semantic and syn-tactic aspects of verbal "eiaboration" andPaivio's (1969) manipulation of the concreteness-image evocativeness of learning materials ex-emplify the "imposed" paradigms to which Irefer. The variations in characteristics ofprose materials by Frase and his associates(e.g., Frase, 1969; Frase & Washington, 1970;Maroon, Washington, & Frase, 1971) are ofparticular relevance here.

A second class of experiments has ex-amined the effect of prelearning instructions,usually in the form of a strategy or mnemonic,which are induced in subjects by experimenters.

2

A technique is introduced by the experimenterwhich, if adopted by the subject, likely willfacilitate the ensuing task. The comparativeeffectiveness of various strategies (notablythose requiring subject-generated verbaliza-tion and imagery) has been studied, summariesof which may be found in the reports of Bower(1v71) and Levin (1971).

In the remainder of this paper, I will dis-tinguish between the imposed and inducedmethodologies which, in fact, I have alreadydone vis-à-vis the Wiener and Cromer (1967)quote. The reader will be helped in makingthis distinction through the use 01 appropriatesection headings, examples, and explicit ref-erences.

Imposed Characteristics andReading Comprehension

Cromer (1970) warted to see if changesin the structure (i.e., organization) of readingmaterials would benefit subjects who had thenecessary identification and vocabulary skills,but who exhibited poor comprehension. As apartial validation of the reading difficultymodels mentioned previously, Cromer selectedsamples of poor-reading junior college studentswith either "deficit" or "difference" problems.The two groups of poor readers were compar-able in mean IQ (Deficit: 110.4; Difference:111.3), but the Deficit group's mean vocabu-lary score (154.4) was lower than that of theDifference group (158.9). The mean vocabu-lary score of the Difference group correspondedto the median score for college freshmen.

Subjects read stories in which the sen-tences wzr-e "organized" in various ways. Intwo of the conditions, the sentences appearedeither in regular form, e.g.:

"The cow jumped over the moon"

or in predetermined phrase groupings, e.g.:

"The cow jumped over the moon"

The latter groupings were based on agreed-upon phrase boundaries as prescribed byLefevre (1964).

The basic finding of the Cromer (1970)study was that when the story was presentedin regular sentence form, there were largedifferences in comprehension between thepoor-reading groups and matched (in IQ) groupsof good readers. However, when the phrasegroupings were employed, the performance ofthe Difference poor readers was as high as

-1h,.? Deficit group's1c,w, in accordance with

That is, a dif--intrtiorl of the matezi.als would be

sur4ects who readothr.,7 than identificationJeticits.

Y ro,=ini?ation of printed ma-... 14-chnique which was found to

orice of poor readers-?nothe-C question which may

l_oor readers comprehendnetter whe:1 they are presented

7 in auditory or pic-printed, form). In a study

Ind Cromer (1971). forthat poor readers corn-

!:(ire. when they listenedv.her, they read them themselves.

eeeee . the reverse was true.7!.: 11,!-Iwer (1971) used two groups

trnm a white middle-class:-ther from a black lower-

1'7 71 Me two groups differed7!: '7111C: presumably) in read-

ere. 7.,tory passages were readel [he rorf,pany of either regular

eeeo ee. or ine drawings which ap--e o s each sentence of the

. er the better readers (middle-make much of a dif-

rited or pictorial accom-: was employed. Per-

: under both conditions. For.e: :lewer-class blacks), when

-1:3mpaniec: by printed sen-e Wc1S very poor; when pic-

were used, performance- that of the better readers.

:-hildren cannot comprehend!e-. quite different from saying

.-cIrnprehend stories.of argument is used by

! 7 5' :11 scUssion of the differ-.. betNNeen Leve'. I and Level II

:!1,:le-and lower-class papule--.relation between IQ (a Level II

learning ability (a Level I mea-e.,';.:,theslzed to be quite strong foree: l-eldren and quite weak for

.±ildren. An implication of this- e s s children with low IQs

learners as xe,..'ell; the samet be made with confidence for

- h:i,lrer, with low IQs, however.f:7 o both the Cromer (1970)

kohwer (1971) studies mayny the all-purpose model iniearning or comprehension is

!t.

assessed on the basis of a single method ormanipulation, only the subjects from Popula-tion I will demonstrate mastery. When a dif-ferent method or manipulation is imposed, themean performance difference between the twopopulations virtually disappears.

This, ot course, gets et the heart of theaptitude-treetment interaction (ATI) debates,the one difference here being that Method Bis better than Method A for both populations,

t comparatively less so for subjects in Popu-lation I. Till now, the ATI'ers have been con-cerned mainly with the detection of disordinal(cross-over) interactions, such that Method Aworks best for Population I and Method Bworks best for Population I. This iype of ATIhas not been easy to demonstrate, howevez(Brecht, 1970; Cronbach & Snow, 1969). Neitherhave aptitudes and treatments in the contextof task and ecological variations been seri-ously considered (Levin, 1971; Salomon, 1971).

An ATI stance is certainly applicable tothe ordinal interaction model portrayed inFigure 1. It simply says that variations in thenature of instructional materials are not ascrucial for "good" students as they are for"poor" students. This should not imply thatvariations in the quality of the organizationor mode of the materials will have no effecton the performance of good students [see, forexample, Maroon et al., 1971]. In the Oakenet al. (1971) experiment already cited, goodreaders comprehended substantially less whenthey read stories which were transcriptions ofpoor readers' efforts to read the normal pas-sage. These transcribed stories included"...all of the poor reader's pauses, falsestr I 's , errors, mispronunciations, omissions ,ete.' Furthermore, as noted earlier, when thegood readers were tested for comprehensionafter listening to stories read to them, theirperformance was worse than when they readthe stories themselves.

An explanaiion provided by the authors isthat in reading the passages themselves, goodreaders are able to go back and re-read anymisunderstood parts of the sentence which, ofcourse, is not possible when listening to asingle spoken version of the same story. Itis also probable that when good readers read,they are employing well-developed organiza-tional strategies which are conducive to com-prehension, and as the authors suggest:

...if poor readers typically do notorganize their input into certain ef-ficacious patterns, they may have con-siderable difficulty understanding whatthey read ....(Oakan et al., 1971, p. 77)

16

3

High

Low

A

Imposed Independent Variable

Figure 1. Hypothesized Ordinal Interaction Between Aptitude (Population)and Treatment (Imposed Independent Variable) to Account forSome Recent Findings

There are two major recommendationswhich follow from the ordinal ATI model inrigum 1. The first of these is that it is in-cumbent upon us to find the optimal presen-tations of learnin materials for children whoappear to be slow learners. Given a sufficientvariety of presentations, many children tabbedas "nonlearners" will emerge as "learners."At the same time, others will hot. It is onlythis !atter group whom we may legitimatelyregard as " nonlearners" and who will re-quire greater remediation than merely changesin materials.

Matz and Rohwer (1c171) have demon-strated that children from middle and lowersocial class groups differ only slightly incomprehension when pictures accompany anauditory version of a story. We have justcompleted a similar study with fourth graderswhich suggests that good and pcor readersfrom the same social class differ less (in

4

terms of comprehension) when a pictorialrepresentation of text is used in place of thetext itself.

If both good and poor readers can com-prehend stories based on pictorial representa-tions, then it is reasonable to ask whethergood reade1.3 are doing something pictorial-like when they are reading regular printedmaterials, while poor readers are not. This

brings us to the second (and potentially moreimportant) recommendation indicated by themodel in Figure I .

Induced Characteristics andReading Comprehension

Notice that until now we hew% -oeen dis-cussing changes in learning materials whichimprove the performance of poo: readers. Thatis, we have considered imposed characteristics

of reading materials. Our attention will nowbe directed toward presumed dii"ferenc.:is inthe t:_sual reading habits of good and poorreaders, If certain of these can be ±denti fled,then it might prove fruitful to instruct or in-L:ut por readers to employ the habits etgood readers when less than "optimally struc-tured" re:ading niaterials are presented. (%ctethat for school- and real world-related read-ing activities, this will generaily be the rulerather than the exception.)

Just what might some cf the habits ofgood readers be ? Certainly these which coe-taie facilitating structures. That self- gener izedvisual imagery is ae established tactie el pro-.ficient r.eaders has been first mildly, and thenstrongly, suggested in a pair of recent stediesby Richard Anderson and his aesociates at theUniversity of :11inois.

The first experiment cAnderson & Hldde,1971) extends some recent findings deal-ing with sentence comprehenston. Bobrow andsower (1969), for example. found that subjectswho were asked either to "disambiguate" (de-termine the contextual meaning of a multiple-meaning word) or to "continue" iconstruct alogical consequence of a prov ded sentencea list of sentences, exhibited recall superiorto subjects who were asked to peruse the earnesentences for spelling errors. Similarly, Beggand Paivio (1969) reported that subjects werebetter able to detect semantic changeeasopposed to lexical changes which had littleeffect on meaning----in a repeated list of sen-tences, especially when the matedals wererelatively concrete. The importance 'Of 'mean-ing" in comprehension has been demonstratedin these and similar experiments (e.g., Bobrow,1970; Levin & Horvitz, 1971). By the sametoken, the role of "imagery" cannot be dis-coented.

In the Anderson and Hicide (1971)study, college students were asked to rateeither the pronunciability or imagery vivid-ness of 30 sentences. On a surprise test forrecall of as many sentences as the subjectcould remember, it was found that the imagery-rating group recalled far more sentences andsentence parts (e.e., verbs and objects) thandid the pronunciability-rating group.

This is interesting in light of the factthat subjects in the latter group actuallyread each sentence aloud (three times) inratine its pronunciability while those in theformer group did not.. On the other hand, thiskind of sentence rehearsal may have beeninterfering (rather than imagery being facili-tative, as has been shown to be the case inintentional learning paradigms with older sub-

jects (e.g., i3±1).n k. Rohwer, GobrowBower, I r-q-° , Within th.7 :rnagerr-ratin gveuc,a moderate relationship was toured between:dojects' reported vividness of a eentence.'sirriagE_Ty and its probability of being recailee.

Ant!ersen's secerei study troveles mradirect evidence for the imagery-reading cors-prehensior. hypothesis. In that experimentAnclerson Kulhavy, )971 high school

seniors were given a written passage to readeither with cr without instructions to visualizevhat tiray were seeding. Although the experi-mental manipulation (i.e., instructions to useimagery) :etas not effective in a "main effect"sense, a pronounced relaticnship was foundbetween subjects' teported frequency cfimagery throughout the paesage and amountof informatior recalled about it (independentlyof their instructional conditions). Thus, thosesubjecte who reported havino used imageryextensively recalled more of what they readthan those who reported having used little orno imagery.

Tn our previously mentioned study withfourth graders, we manipulated the degree towhetes subjects presumed to be non-imageryproducers (poor readers) generated visualin.ages while readine, by inducing imageryin theirs or not. As preciteted, he Imag.5tratecIr Improved comprIthensiz.n, and ina:;cordance with t!-..e Wiener and Cromer (1967)models, "difference" poer readers (those withadequate vocatelary skills) benefitted moretnan "deficit" poor readers (those lackingprerequisite vocabulary skIlls). The Inter-pretation of such results is similar to Cromer's

7C) and, of course, has important ilnplica-tions with regard to the teaching of readingto chltdren who re.ad poorly for diverse reasone.The trainirei of imagery production in childrenin need of an organizational framework looms

a reasonehle strategy.Nhich of what have been saying has been

thought, if riot articulated, by others before(including reading experts). It is especiallynoteworthy that in one of the standard readingtests, among the author's' suggestions for im-proving the reading of low achievers may befound:

1

Word recognition practice, phrase prac-tice and expression practice shoeldrequire response to meaning and im-agery. Reading is getting ideas fromthe printed word; all aspects cf read-ing instruction should focus uponmeanings and reactions to meanings.(Durrell & Hayes, 1969; Durrell &Brassard, 19691

These comments appear on the primary (Grades1- 3.5 \ zwl intermediate (Grades 35-(,) vet-&Ions of the Durrell test, and are apparentlyintended for "deficit" poor readers. However,a genera)ization of the imagery idea fromcomprehension cf words to comprehensionof sentences 3nd paragraphs is not includedin Durrell's (1969) suggestions on the ad-vanced (Grades 7-9) version of his test, wherea greater proportion of poor readers are prob-ably of the "difference" variety.

As educators, we should continually beseeking ways in which the learning processmay be improved and hopefully be made moreenjoyable. Changing characteristics of exist-ing matedals or changing students' charac-teristic learning behaviors seem to be tworeasonable (and not necessarily mutually ex-clusive) possibilities. While the "imposed"

6

4,3

ap;)roach caters to individual and group thf-terences through the presentation of differ-entially effective organizations, the "induced"approach promises the greater educationalpayoff beyond the confines of the well-organized textbook, the optimally sequence,1teaching machine. and the multitalentedteacher. Equipped with efficient inducedstrategies, the child will be less dependenton the ::uality of stimuli in his environment,for he will be capable of reorganizing, ela-borating, and concretizing relatively disor-ganized, unelaborated, and abstract materials.Of course, to anticipate this without regardto the auxiliary efforts required (for example,shaping such behaviors over time with thehelp of appropriate reinforcers) is beyondcomprehension.

References

Anderson, R. C., & Hidde, J. L. Imagery andsentence learning. Journal of Educa-tional Psycholo.i,y, 1971, 62, 526-530.

Anderson, R. C., & Kulhavy, R. W. Imageryand prose learning. Unpublished manu-script, University of Illinois, 1971.

Bean, J.13., & Rohwer, W. D., Jr. Noun pairlearning in adolescents: Population dif-ferences. Paper presented at the annualmeeting of the American Educational Re-search As...ociation, New York, February1971.

Begg, I., & Pelvic, A. Concreteness andimagery in sentence meaning. Journalof Verbal Learning and Verbal Behavior.1969, 8, 821-827.

Bobrow, S. A. Memory for words in sentences.Journal of Verbal Learning and VerbalBe. havior , 1970. 9, 363-372.

Bobrow, S. A., & Bower, G. H. Comprehensionand recall of sentences. Journal of Ex-perimental Psychology, 1969, 80, 455-461.

Bower, G. H. Mental imagery and associativelearning. In L. Greeg Ed.), Cognition inLearning and Memory. New York: JohnWiley & Sons, 1971.

Brac.ht, G. H. Experimental factors related toaptitude-treatment interactions. Reviewof Educational Research, 1970, 40, 627-645.

Cromer, W. The difference inc. ',1: A new ex-planation for some reading difficulties.Journal of Educational Psychology, 1970,61, 471-483.

Cronbach, L. T., & Snow, R. C. Individual dif-ferences in learning ability as a functionof instructional variables. Final report.USOE, Contract No. ()EC-4-6-061269-:217, March 1969.

Davidson, R. E. Semi-grammaticalness in thefree learning of sentences. Unpublisheddoctoral dissertation, University of Cali-fornia, Berkeley, 1966.

Durrell, D. D. Durrell Listening-ReadingSeries, Advanced Level. New York:Harcourt, Brace & World, 1969.

Durrell, D. D., & Brassard, M. B. DurrellListening-Reading Series. IntermediateLevel. New York: Harcourt, Brace &World, 1969.

Durrell, D. D., & Hayes, M. T. DurrellListening-Reading Series, PrimaryLevel. New York: Harcourt, Brace &World, 1969.

Frase, L. T. Paragraph organization of writtenmaterials: The influcence of conceptualclustering upon the level and organiza-tion ef recall. Journal 0- EducationalPsychology, 1969, 60, 394-401.

Frase, L. T., & Washington, E. D. Children'sability to comprehend text. Paper pre-sented at the annual meeting of the Amer-ican Psychological Association, Miami,September 1970.

Jensen, A. R. How much can we boost IQ andscholastic achievement? Harvard Edu-cational Review, 1969, 39, 1-123.

Lefevre, C. A. Linguistics and the Teachingof Reading. New York: McGraw-Hill,1964.

Levin, J. R. Issues in imagery and learning:Verbal and visual variables. Symposiumpaper presented at the annual meeting ofthe Western Psychological Association,San Francisco, Aglril 1971.

Levin, J. R., & Horvit4t, J. M. The meaning ofpaired associates,. Journal of Educa-tional Psychologi, 1971, 62, 209-214.

Levin, J. R.. Rohwer, . D., Jr., & Cleary, T. A.Individual differences in the learning ofverbally and pictorially presented pairedassociates. American Educational Re-search Jou741a1, 1971, 8, 11-26.

Maroon, S., Washington, E. D., & Prase. L. T.Text organization and its relationshipto children's comprehension. Paper pre-sented at the annual meeting of the

7

American Educational Research Associa-tion, New York. March 1971.

Matz, R., & Rohwer, W. D.,Jr. Visual elabora-tion of text. Paper presented at the an-nual meeting of the American EducationalResearch Association, Now York, March1971.

Oakan, R., Wiener, M., & Cromer, W. Identi-fication, organization, and reading com-prehension for good and poor readers.Journal of Eth.cational Psychology, 1971,62, 71-78.

Paivio, A. Mental imagery in associativelearning and memory. PsychologicalRericu , 1969, 76, 241-263.

Paivio, A. Imagery and language. ResearchBulletin No. 167, University of WesternOntario. 1970.

,op

8

Rohwer, W. D., Jr. Social class differencesin the role of linguistic structures inpaired-associate learning: Elaborationand learning proficiency. Final report,USOE, Contract No. OE 6-10-273, No-vember 1967.

R hwer, W. D., Jr. Images and pictures inchildren's learning: Research resultsand educational implications. Psycho-logical Bulletin, 1970, 73, 393-403.

Salomon, G. Aptitude-treatment interactions.Colloquium held at the annual meetingof the American Educational ResearchAssociation, New York, March 1971.

Wiener, M., & Cromer, W. Reading and read-ing difficulty: A conceptual analysis.Harvard Educational iik-Vicit' , 1967, 37,620-643.

GPO 11116-184-2