Students' learning difficulties in science : causes and possible remedies

INVESTIGACION EXPERIENCIAS DIDACTICAS STUDENTS' LEARNING DIFFICULTIES IN SCIENCE. CAUSES AND POSSIBLE REMEDIES KEMPA, R. F. Uni e si y o Keele, S a o dshi e, UK. SUMMARY In his pape , a b oad analysis is p esen ed o he main ypes o lea ning di icul ies encoun e ed by s uden s in science educa ion and causes unde lying hem. Teache s and educa ionis s make equen use o he e ms 'lea ning p oblems' and 'lea ning di icul ies', bu no gene ally accep ed de ini ion exis s o hei p e- cise meaning. Ye , i seems o me ha wi hou some cla i ica ion o hese e ms i is ha dly possible o begin o iden i y such di icul ies, le alone alk meaning ully abou hei na u e and abou how hey can be o e come o emedied. I is possible, o cou se, o apply he e m 'lea ning di icul y' o any si ua ion whe e a s uden ails o unde s and a ela i ely easy concep o no ion ha we wish him o he o acqui e as he esul o some ins uc- ional in e en ion. Such desc ip ion o 'lea ning di i- cul y' is o li le p ac ica1 alue, hough: i is oo gene al and all-emb acing and, in addi ion, does no e en hin a he possible causes o such di icul ies. Le me, he e- ENSEÑANZA DE LAS CIENCIAS, 1991,9 (2), 119-128 119 o e, p opose a somewha na owe de ini ion: "A leaming di icul y nay be said o exis in any si ua- ion whe e a s uden ails o g asp a concep o idea as he esul o one o mo e o he ollowing ac o s: i. The na u e o he ideas/knowledge sys em al eady possessed by he s uden , o he inadequacy o such knowledge in ela ion o he concep o be acqui ed. ii. The demand and complexi y o a lea ning ask in e ms o in o ma ion p ocessing, compa ed wi h he s uden 's in o ma ion-handling capaci y. iii. Communica ion p oblems a ising om language use, e.g., in ela ion o echnical e ms o o gene al e ms wi h con ex -speci ic specialized meanings, o he com- plexi y o sen ence s uc u e and syn ax used by he eache (compa ed wi h he s uden 's own language capaci y). i . A misma ch be ween ins uc ional app oaches used by he eache and he s uden 's p e e ed lea ning mode (lea ning s yle)." No claim is made o his de ini ion o be comp ehen- si e, and he e nay well be o he poin s ha could and should be added o i . Wha he de ini ion does a emp , howe e , is o d aw a en ion o a ange o causes o lea ning di icul ies which can possibly be emedied h ough app op ia e pedagogical in e en ions o cu icula s a- egies. I ough o be acknowledged a his s age ha a s uden 's ailu e o lea n wha we wish him o he o lea n, canno always be a ibu ed o 'lea ning di icul ies' in he sense o he abo e de ini ion. O he easons nay also p e ail and be esponsible o s uden s' ailu e o lea n, e.g., lack o in e es , an inadequacy o e o o a lack o a en ion. P ac ising eache s a e well awa e o he impo an e o hese ac o s-indeed, hey equen ly in oke hem in o de o 'explain' lack o pe o mance and low achie emen on hei s uden s' pa . Despi e hei ob ious impac on s uden s' lea ning, 1 iew hese ac- o s as ela ing o wha Gagné (1965) has desc ibed as "p econdi ions o lea ning", a he han o he leaming p ocess i sel . Fo his eason, 1 ha e excluded hem om my lis o causes o leaming di icul ies. The di ision o s uden s' lea ning di icul ies and hei unde lying causes in o he abo e ou a eas is, o some ex en , a ma e o con enience and expediency. I e lec s he ac ha hese ca ego ies co espond o ai ly well-es ablished a eas o esea ch. This is no o say ha he a eas should be iewed as ela i ely independen o one ano he . On he con a y, he e a e de ini e links be ween hem. Bo h he i s and he second a ea indica ed abo e nay be said o ela e o he lea ne s' knowledge and memo y s uc u e, bu each does so in a di e en way. Whe eas he i s is conce ned wi h he na u e o he knowledge held by lea ne s in wha , in e ms o an in o ma ion- p ocessing model, is gene ally e e ed o as hei 'long- e m memo y' (LTM), he second a ea ocuses on s u- den s' 'sho - e m memo y' (STM), some imes also e- e ed o as 'wo king memo y'. When conside ing inadequacies in wo king memo y as a sou ce o lea ning di icul ies, he ocus is no on he s uc u e o ideas held by s uden s in hei memo y, bu on he capaci y o s uden s o hold and manipula e di e en pieces o in o ma ion simul aneously. Con- side able expe imen al e idence exis s now e sugges ha people di e wi h espec o he numbe o i ems o in o ma ion on which hey can ope a e in asks equi ing he ans o ma ion o in o ma ion (lea ning asks in a- iably equi e his). The e o e, any 'misma ch' be ween he complexi y o a lea ning ask (when seen in e ms o he numbe o in o ma ion i ems o be handled) and a s uden 's in o ma ion-handling capaci y mus be seen as a po en ial sou ce o lea ning di icul ies. The hi d a ea sugges ed abo e co e s a ange o issues ha nay be summa ized unde he label 'language and communica ion'('). This is a widely esea ched a ea, pa icula ly in ela ion o he explo a ion o s uden s' unde s anding o echnical and non- echnical wo ds in science. P ominen among hese s udies in he English- speaking con ex a e hose by Ga dne (1972) in Aus a- lia and by Cassels and Johns one (1977) in B i ain. In he Spanish con ex , he wo k o Llopis and colleagues a Valencia is no ewo hy. The pa icula aspec s o 'language and communica ion' which cause o nay con ibu e o lea ning di icul ies in science, a e oo nume ous o be lis ed he e. Le me me ely indica e ha hey ex end beyond s uden s' unde s anding o he meaning o wo ds. Fac o s such as he complexi y o eache s' speech and/o ex ual ma e- ials o use by s uden s a e p obably no less impo an in his con ex , especially when one ecognizes ha 'complexi y ' is a mul i- ace ed concep . O he language- ela ed aspec s o conside able in e es in science edu- ca ion include he use o me apho s and he e ms and analogies used in he desc ip ion o abs ac no ions and concep s. The ou h and inal a ea emb aced by he abo e de ini- ion o lea ning di icul ies conce ns he ela ionship be ween eaching s yle (as exempli ied by he ins uc- ional app oach adop ed by a eache o ecommended o he implemen a ion o a cu iculum) and lea ning s yle (which nay be hough o as ep esen ing a s uden 's 'na u al' o p e e ed mode o a ending o a lea ning ask). F equen e e ence has been made in he li e a u e o he desi abili y o ma ching eaching s yles o s u- den s' leaming. Resea ch e idence in suppo o his no ion is, howe e , a he scan . Compa ed wi h he i s h ee a eas p oposed abo e, he ou h is a he less well de ined concep ually and mus , hence, be mo e specula i e in e ms o i s p ac ica1 explo a ion. No ions such as eaching s yle and lea ning s yle a e exceedingly complex. Excep o some ela- i ely coa se a emp s, we s ill lack a good ope a ional iden i ica ion and cha ac e iza ion o eaching s yles. (Eggles on and co-wo ke s, in 1976, a i ed a h ee b oad ' ypes' o science eache , each cha ac e ized by i s own eaching s yle.) The ask o concep ualizing and iden i ying genuine lea ning s yles is no less di icul . Ho s ein and Kempa (1985), ollowing a sugges ion by Ada (1969), made a limi ed a emp in his di ec ion in ela ion o science lea ning. This was based on he p emise ha s uden s' lea ning s yles a e e lec ed in hei p e e ences o , s dislike o , pa icula ins uc ional p ocedu es. Recen ly, Kempa and Ma in (1989) ha e explo ed his a ea empi ically. (See also Kempa and Ma ín Díaz 1990 a, b.) The ou a eas sugges ed he e di e conside ably in e ms o he amoun o a en ion ha has been gi en o hem by science educa ion esea che s. The i s a ea, in pa icula , has been he subjec o conside ably mo e esea ch han any o he o he a eas. Hence, ou insigh -as science educa o s and eache s-in o he ela ionship be ween s uden s' lea ning (and p oblem-sol ing) di i- cul ies and hei long- e m memo y is now ai ly weBl de eloped. In ela ion o he o he a eas, ou knowledge base is somewha weake . (This is especially ue o he ou h a ea men ioned, which conce ns he ma ching o each- ing app oaches and lea ning s yles.) This means ha he ex en o which esea ch indings in hese a eas a e ansla able in o guidelines and sugges ions o cu icu- la ac ions is limi ed. E en i such guidelines a e de eloped, hey may ha e o be iewed wi h some cau ion, as a as hei alidi y and gene alizabili y a c conce ned. Howe e , his does no make hese a eas less impo an han he i s one. Wha i does mean is ha in u u e inc eased a en ion needs o be gi en by science eache s and esea che s o hese a eas also. EXPLORING THE IMPLICATIONS OF RE- SEARCH INTO LEARNING DIFFICULTIES -THE ROLE OF MODELS OF LEARNINGl TEACHING The majo idea unde lying he whole a ea o lea ning di icul ies, a leas in ela ion o he causal ca ego ies discussed in he p eceding sec ion, is ha o 'misma ch'. By his 1 mean ha lea ning di icul ies do no exis iiz acuo, bu appea only in si ua ions whe e eaching in en isns and in e en ions a e no adequa ely ma ched o he lea ne 's in ellec ual capabili ies o cogni i e unc ioning. The ac ual na u e o he misma ch can, o cou se, a y conside ably (as has al eady been explained in he p e ious sec ion), bu gene ally misma ch is a - ibu able o some 'de iciency ' in he lea ne 's knowledge s uc u e o hislhe e bal epe oi e, e c. F om a heo e ical pe spec i e, any a emp o emedy o educe a lea ning di icul y mus add ess i sel o he nega ion o he unde lying misma ch o de iciency, by means o app op ia e in e en ions. In p incipie, such in e en ions can ake wo o ms, iz., - he p o ision o supplemen a y ins uc ion designed o ill knowledge gaps andlo widen he meanings ha , s uden s associa e wi h pa icula no ions and concep s; - he modi ica ion o ins uc ional app oaches ei he o a oid o by-pass he pa icula causes o a lea ning di icul y, o o p o ide he lea ne wi h expe iences ha enable him/he o esol e he di icul y. I is p obably no un easonable o assume ha a eache o cu iculum de elope , in adop ing a pa icula mode o in e en ion, does so in he expec a ion ha i is 'app op ia e' o a pe cei ed lea ning di icul y, i.e., ha i will emo e o esol e he la e . The only excep ion he e would occu i an amelio a ion s a egy we e chos- en en i ely on a ial-and-e o basis. Igno ing his excep ion, we may claim ha in e en ions chosen by eache s o o e come pa icula lea ning di icul ies e lec hei assump ions no only abou he causes o he di icul ies, bu also abou he me hods sui able o esol ing hem. Such assump ions a e essen ially he esul o 'model cons uc s' o leaming and eaching, i espec i e o whe he hese a e held on an indi idual basis o whe he hey enjoy consensus s a us. I ollows om he o egoing ha he concep ion o lea ning and eaching held by a eache o a eu iculum de elope is likely o be a majo de e minan o he s a egy chosen by him/he o he nega ion o educ ion o lea ning di icul ies. Fo his eason, some conside a- ion has o be gi en o he ole o models o eachingl lea ning in de eloping s a egies o emedying lea ning di icul ies. Fo a lea ning heo y o be po en ially use ul as a basis o he choice o pedagogical in e en ions o o e com- ing leaming di icul ies, i has o ul il wo essen ial equi emen s: i. I mus inco po a e cha ac e is ics ha ela e o add ess hemsel es o he causes o lea ning di icul ies, and ii. i mus in oke o pos ula e some mechanism whe e- by knowledge change, e c., is b ough abou . In ela ion o he i s o hese poin s, he obse a ion has o be made ha no single lea ning heo y ha is cu en ly a ailable, is capable o accoun ing adequa ely o he whole spec um o causes o lea ning di icul ies. The e o e, se e a1 lea ning heo ies ha e o be called upon i we a e o deal wi h di e en ypes o lea ning di icul y. This is no a se ious d awback, p o ided ha he di e en heo ies complemen , a he han con adic , one an- o he . The second equi emen speci ied abo e should be la gely sel -explana o y. I is based on he ecogni ion ha non- mechanis ic lea ning heo ies a e usually 'desc ip i e' in na u e and so ha e li le alue as ounda ions o making decisions abou he choice o pedagogical in e - en ions and s a egies (c . Kempa 1983). Mos o he heo e ical posi ions ha unde lie, o a e implici in, esea ches in o s uden s' lea ning di icul- ies andlo a emp s o emedy hem may be associa ed wi h wo majo pe spec i es. These a e he cons uc i is pe spec i e o lea ning and he in o ma ion-p ocessing BNSEÑANZA DE LAS CIENCIAS, 1991,9 (2) INVESTIGACION Y EXPERIENCIAS DIDACTICAS pe spec i e, espec i ely. Following Osbo ne (1985), he dis inc ion be ween hese pe spec i es is bes exp e- ssed in e ms o hei espec i e conce ns: "A cen al ea u e o he cons uc i is adi ion con- ce ns knowledge: he concep s, meanings, unde s- andings...and in ellec ual skills s o ed by an indi idual in memo y"... [whe eas] " he in o ma ion-p ocessing adi ion is mo e conce ned wi h he p oblem-sol ing capaci y o he human mind and wi h he hinking p o- cesses ha occu when p oblems a e ackled." This dis inc ion is use ul in ha i poin s o he cons uc i is pe spec i e as he model app op ia e o he conside a ion o lea ning di icul ies which a ise om inadequacies in he lea ne 's knowledge s uc u e, i.e., his/he long- e m memo y. In con as , he in o ma ion-p ocessing pe spec i e has ob ious ele ance o lea ning p oblems which s em om di icul ies in in o ma ion-handling and ans o ma ion, which ela e o he lea ne 's wo king memo y. The language a ea (which o ms , he hi d o he ou a eas iden i ied in ou de ini ion) can also be linked o he cons uc i is model, p o ided ha language is e- ga ded as a means o lea ning, a he han jus as a communica ion sys em. The o me iewpoin has ound inc easing accep ance in ecen yea s. To quo e Solo- mon (1987): "The cons uc ion o meaning h ough language... does mo e han ex end and e ine he ocabula y–i builds up he e y pic u e o eali y jus as powe ully as does he sum o senso y pe cep ions and expe iences." Summa izing he discussion in his sec ion so a , we may claim ha he cons uc i is heo y o lea ning and he in o ma ion-p ocessing model, be ween hem, o m a easonable heo e ical basis o he in e p e a ion o he i s h ee a eas o lea ning di icul ies iden i ied in he p eceding sec ion. Mo eo e , i is possible o de i e om hese heo e ical models a se ies o pedagogical p inciples o a oiding and/o emedying lea ning di i- cul ies. As has al eady been s a ed, in ela ion o he ou h a ea o lea ning di icul ies (i.e., ha conce ned wi h he misma ch be ween lea ning s yles and ins uc ional s yles) ou heo e ical unde s anding is less well de eloped. So is, in consequence, ou abili y o a icula e pedagogical s a egies whe eby his ype o misma ch can be a oided o educed. The eason o his si ua ion lies la gely in he complexi y o he lea ning/ eaching p ocess. Al- hough nume ous esea ches ha e explo ed speci ic as- pec s o lea ning/ eaching in e ac ions, in o ma ion de- i ed om hem is s ill oo agmen ed o allow us o make holis ic p onouncemen s abou he ins uc ional p ocess. Ne e heless, some heo e ical posi ions can be a i ed a which, al hough a om comp ehensi e, may be used as a s a o a close ma ching o eaching and lea ning s yles. Kempa and Ma in's (1989, 1990a, 1990b) s udy o he ela ionship be ween s uden s' mo i a ional pa - e ns (which mani es hemsel es in p e e ences o di e en lea ning modes) and ins uc ional s a egies in science educa ion, o ms an example o his. SOME EXAMPLES AND APPLICATIONS In he discussion so a , he emphasis has been on he cla i ica ion o he concep o 'lea ning di icul ies in science' and on he es ablishmen o some heo e ical amewo k wi hin which such lea ning di icul ies can be unde s ood, in e p e ed and possibly emedied. My in en ion now is o illus a e some o he aspec s men ioned abo e by e e ence o some ac ual examples and esea - ch indings. Ha ing so s ongly ocused on he ole o long- e m memo y and wo king memo y as key ac o s in causing lea ning di icul ies, I p opose o pay pa icula a en ion o hese wo aspec s, o he exclusion o he o he s. I may be help ul a his junc u e o e lec b ie ly on he espec i e unc ions o he wo memo ies in lea ning and/o p oblem-sol ing. Figu e 1 helps o illus a e hese in a schema ic o m. We may hink o he long- e m memo y as a as knowledge s o e om which in o ma ion is d awn and o which in o ma ion is added om ime o ime. The wo king memo y in his ep esen a ion may be likened o he cen al p ocessing uni o a compu e sys em: i is he e whe e he in o ma ion de i ed om he long- e m memo y and om ou side is b ough oge he in men al ope a ions and ans o ma ions. The esul s o hese p ocesses mus ine i ably e lec (i) he na u e o he in o ma ion held in he lea ne 's long- e m memo y and (ii) he capaci y o his/he wo king memo y –which ob iously links o he poin s p e iously made. Figu e 1 Schema ic ep esen a ion o in o ma ion-p ocessing model. INFORMATION INPUT FROM OUTSIDE ---> LONG-TERM MEMORY STORE SHORT-TERM (WORKING) MEMORY RESPONSE TO LEARNING TASK OR PROBLEMS 1. Long- e m memo y s uc u e and science lea ning The li e a u e abounds wi h epo s on he iews and ideas on scien i ic phenomena, held by pupils and s u- den s be o e and a e ins uc ion. The las ew yea s in pa icula ha e seen a p oli e a ion o s udies in his a ea. As a esul , we ha e now a ailable a good deal o in o ma ion abou many o he ideas and misconcep ions held by pupils and s uden s abou scien i ic phenomena and e en s [see, e.g., D i e and E ickson (1983) and Gilbe and Wa s (1983)]. 122  ENSE&ANZA DE LAS CIENCIAS, 1991, 9 (2) Le me gi e he e jus one example o indings abou s uden s' unde s anding o scien i ic phenomena, in a de o explo e he ela ionship be ween long- e m memo y s uc u e and science lea ning. The example is aken om he Child en's Lea ning in Science p ojec (CLIS 1987), ecen ly conduc ed in England. The p ojec se ou o iden i y in a sys ema ic way he ideas held by seconda y school s uden s abou a ange o scien i ic hemes and o explo e ways o emedying any misconcep ions. One o he hemes in es iga ed conce ned he ela ion- ship be ween he beha iou o gases and hei pa icle na u e. S uden s' unde s anding o his was examined by ques ions such as his: "A e many expe imen s, scien is s now hink ha - al1 hings a e made o small pa icles - hese pa icles mo e in al1 di ec ions - empe a u e a ec s he speed hey mo e - hey exe o ces on each o he . Use any o hese ideas o explain he ollowing: A oo ball is pumped ha d du ing he day when i is wa i. In he e ening when he empe a u e alls, he oo ball does no eel so ha d." To illus a e he ange o no ions held by s uden s abou his phenomenon, le me gi e a ew o he ac ual answe s ecei ed. Answe a: , "When i is wa m, he ai pa icles in he oo ball a e mo ing a ound as and a e sp ead ou . Howe e , when he oo ball becomes coole , he ai pa icles slow down and come nea e o each o he , Le., he ai con ac s. The e o e he e is less p essu e on he ball ..." Answe b: "When he wea he is wa m, he pa icles inside he ball ha e expanded so ha he ball will hold no mo e ai ... In he e ening when he wea he is no so ho , he pa icles sh ink so hey no longe ill he space in he ball ..." Answe c: "When i is wa m, he pa icles in he ai (inside he bal'l) a e mo ing al1 o e and so make i ha d ... Bu in he e ening, when he empe a u e d ops, he pa icles s op mo ing abou ..." These s uden 's answe s, which a e ypical o a wide ange o answe s ecei ed, e eal ce ain no ewo hy cha ac e is ics. Answe a is clea ly accep able in he scien i ic sense. I b ings ou he link be ween empe a u e and he speed a which he pa icles in he ai mo e, which is he essen ial ea u e o a co ec in e p e a ion o he phenomenon. Answe s b and c, in con as o answe a, exp ess dis inc misconcep ions. The i s o hese, appea ing in b, is h;i he pa icles hemsel es unde go a change in size as he empe a u e o he ai changes (' he pa icles expand', ' he pa icles sh ink'). The second, shown in c, s ipu- la es ha he pa icles a e no necessa ily in con inuous mo ion ( hey 's op' on cooling). Misconcep ions like he o egoing (and he e a e many o he s in e idence in o he a eas o science) a e eadily a ibu able o 'inadequacies' in s uden s' long- e m memo y s uc u e, as al eady hin ed a ea lie , In he case o he i s misconcep ion, be s uden e iden ly made an inco ec connec ion be ween he mac oscopic p ope ies o a bulk o ma e and he mic oscopic pa icles o which ma e is composed. (Ano he exam- ple o his, equen ly encoun e ed in science eaching, is s uden s' assump ion ha he a oms o a me al such as coppe o gold, ha e he same colou as he me al i sel .) The second misconcep ion likewise can be in e p e ed in e ms o inapp op ia e links be ween knowledge i ems in he lea ne 's memo y s uc u e. The wo knowledge i ems in ol ed he e a e ha a. he educ ion in empe a u e causes a slowdown o he pa icles, and b. a slowdown in he speed o an objec esul s ul ima e- ly in he loss o any mo emen . The la e is, o cou se, a common e e yday expe ience associa ed wi h, o example, he b aking ac ion o a ca . The kinds o link be ween knowledge elemen s in a s uden 's memo y s uc u e ha we a e s ipula ing he e, mus , in he i s ins ance, be ega ded as hypo he ical cons uc s. The ques ions a ise whe he such memo y s uc u es may ac ually be un olded and, i so, whe he hey gi e any clues abou s uden s' lea ning di icul ies, Some o oii own wo k a he Uni e si y o Keele has sough o answe hese ques ions. Using a echniqiie called wo d-associa ion es ing, we ha e explo ed s u- den s' knowledge s uc u es in ela ion o selec ed a eas o chemis y and p esen ed hese in he o m o 'cogni- i e maps'. Compa ing he cogni i e maps o s uden s o di e en abili y in he a eas co e ed by he maps has enabled us o shed some ligh on he in luence o s u- den s' knowledge s uc u e on hei lea ning and p o- blem-sol ing beha iou . In hei simples o m, cogni i e maps show g aphically he links and associa ions among di e en ela ed con- cep s and no ions, as hey exis in a s uden 's mind. I is no necessa y o he na u e o he links o be nade explici , as long as one can easonably in e p e he cogni i e maps. Le me illus a e his by wo e~amples which ela e o he concep a eas 'acid' and 'elec oly- sis', espec i ely. In each case, we con as he cogni i e maps o s uden s who, in p oblem-sol ing asks ela ing o he wo concep a eas, had been success ul and unsuc- cess ul, espec i ely. The cogni i e maps ep esen a- ions a e gi en in igu es 2 and 3. Conside ing igu e 2 i s , i is seen ha he main di e - ence be ween he wo cogni i e maps lies in he absence ENSENANZA DE LAS CIENCIAS, 1991,9 (2) Figu a 2 "A e age" cogni i e maps o he concep ACID, o (a) good and (b) poo p oblem-sol e s. om ha o he 'poo ' p oblem-ssl e s o a se o concep s/no ions which may be said o be 'abs ac ' in na u e. In al1 o he de ails, he maps appea iden ical. Wha we may conclude om hese maps, is ha s uden s who, in ela ion o ma e s conce ning 'acids', u n ou o be poo p oblem-sol e s, lack pa icula knowledge elemen s ha a e usually ega ded as impo an : in his case, he 'absen ' concep s a e he p o on, o hyd ogen ion (Kempa and Nicholls 1983). Figu e 3 which is based on expe imen al wo k by Slim- ming (1984), is signi ican ly mo e complex in ha i deals wi h a b oade subjec a ea (elec olysis). Compa ing he wo ep esen a ions, i is no ed ha he knowledge elemen s ound in he wo maps a e app oxima ely he same. Nowe e , he ep esen a ions di e subs an ially in he na u e and ex en o he links be ween concep s and knowledge elemen s. Figu e 3(a) shows links and associa ions which, o he expe ienced science/chemis y eache , appea logical and jus i iable. Figu e 3(b), in con as , con ains ewe connec ions ( he cogni i e map is mo e agmen ed) and qui e a ew o hese a e illogical, om a scien i ic poin o iew. We should, o cou se, no be su p ised o ind ha he cogni i e s uc u es, as hey exis in s uden s' long- e m memo ies, ha e conside able in luence on he way in which s uden s can g asp o make sense o new in o ma ion and ideas. Thus, a di ec connec ion eme ges be ween cogni i e s uc u e (long- e m memo y s uc u e) and lea ning di icul ies. Ou own indings and hose o o he esea che s lead o he conelusion ha long- e m memo y ela ed lea ning di icul iea a e usually a ibu able o one o mo e o he ollowing ac o s: - The absenee o knowledge elemen s om a s uden 's memo y s uc u e. Figu a 3 Cogni i e maps o he concep Ele olysis. The op nap applies s a good p oblem-sol e , whils he bo om map is ha o a poo p oblem- ol e . - The exis ence, in he s uden 's memo y s uc u e, o w ong o inapp op ia e links and ela ionships be ween knowledge elemen s. - The absence o essen ial links be ween knowledge elemen s in he s uden " memo y s uc u e. - The p esence o alse o i ele an knowledge ele- men s in he s uden 's memo y s uc u e. Each o hese 'de iciencies' is illus a ed by he cogni- i e maps p esen ed he e. These maps deal wi h ela i e- ly small a eas o subjec ma e . Fo mo e complex ields, mo e ex ensi e maps could p obably be cons ue- ed, bu hei inc easing complexi y would make hem mo e di icul o p esen g aphically and o in e p e . 1 ha e no so a e e ed o lea ning p oblems ha a ise om pupils and s uden s holding scien i ic no ions ha a e in con lic wi h hose accep ed by scien is s. These a e o en e e ed o in he li e a u e as al e na i e ia ne- wo ks, child en's science o commonsense science, e c. Essen ially, hese a e co e ed by he p esen discussion also: al e na i e amewo ks may be iewed as he esul o s uden s holding knowledge s uc u es, ha a e di e en om hose o he scien is . The ypes o de iciency in (long- e m) memo y s uc u e p o ide a leas a pa ial explana ion o he lea ning di icul ies ha we equen ly de ec ou s uden s o ENSEÑANZA DE LAS CIENCIAS, 1991,9 (2) ha e. The ecogni ion and diagnosis o hese de i- ciencies also p o ides a na u al s a ing poin o he design o eaching s a egies by means o which s uden s' lea ning di icul ies can be educed o o e come. One such eaching s a egy is ha de eloped and used by he 'Child en's Lea ning in Science p ojec ' (CLIS) men ioned ea lie (Needham and Hill 1987). I is based on he cons uc i is model o lea ning and inco po a es ea u es p e iously sugges ed by au ho s such as Nuss- baum and No ick (1982) and No ak and Gowin (1985). Table 1 ou lines he main cha ac e is ics o his eachiig app oach. As can be seen, g ea emphasis is placed on i. he p o ision o oppo uni y o pupils o beco ne awa e o he ideas which hey hold p io o being augh (Elici a ion o Ideas); ii. he modi ica ion, ex ension o eplacemen o such ideas, i ound inadequa e, by new ones (Res uc u ing o Ideas). The eaching sequence implici in able 1 has been aiily success ully applied by he CLIS eam in se e a1 a eas o science eaching. Table 1 Teaching sequence p oposed and used by he CLIS p ojec (c . Needham and Hill 1987). 2. Wo king memo y capaci y and lea ning di icul ies PIUSO L ORIEMA~ON R EllC AnON 3F IDP. 11L RESTRUC7 jRMO OF IDUS 1 Clanñalhn id nclung 2 F-u e u> mn l ú iiwilo<ii 3 Couin nwn o1 ne idu 4. Enlulilon IV APPLlCATiON OF IDEAS V REVIEW In he in oduc o y sec ion o his pape , a en ion was d awn o he cen al ole played in lea ning p ocesses by he sho - e m o wo king memo y (c . igu e 1). We may in e p e 'wo king memo y' as ha pa o a pe son's in ellec ual unc ioning in which in o ma ion om ex- e na1 e en s and om one's long- e m memo y is b ough oge he and p ocessed. I is well es ablished h oughpsychological esea ch ha he capaci y o ou wo king memo y is a he limi ed. Fo example, acco ding o Mille 's s udies in he 1950s (Mille 1956), mos people can hold only abou 7IT2 in o ma ion uni s in hei sho - e m (wo king) memo y. B oadben (1975) e en sugges s ha Mille 's igu e is oo high; bu mo e impo an in he p esen con ex han a deba e o whe he he a e age igu e is 7 o less a e he ollowing indings om educa ional esea ch: PL RHISL &o- m c ai aod u< <h Y~DL To embk pupO iad d 10 bcmn m e o p b W To c aie an an eac o llemlhc inpou - lk wunilllc aoe - <o. 2 Za c eplu m h i mo e wnuñc nn Recugik alie nalm la lad c l i lg enmine ap To nlidli)i o cn lq i4c ModiIy. mend o cph miing idea Tal ~ilidi y o1 nmiy mmw cd id- Reinlo cmcn o1 mm<n íd !den 10 hmilu 1d aud I ul IO~ Wne > O chanp O1 idc and hmilimn iiion nlb lumia' pmce lo llai ibc pupa o e lm "pon ibc mcnl 10 W ihcu id- h- chang& - The wo king memo y capaci y is di e en om pe son o pe son. W XHODS P c ilmi>c* -1 pmbkm 10 sol c. wha Jcmominibm. ñim Almi* i@ nmppn culllnm Ruilal m wii i a SMU pup dluiuswn, Collo ed b>. cponlng bu Small pup du~u sion and epon ng bac Tuche dc non i a ion. p pil e pcnmenu. wckshau DL-ioa udnnb luche? inpu Pn iid w , p ola w . expcnmenlaiion. eache ~~~ONI~IIO& Panoiul amllnb p aaial anhi)i. p oblem uilwag. pmw w k kMNl wniinb @Oup d-10% pmnal dime. oieuiing w k po en eic - Fo a pa icula pe son, he wo king memo y capaci y appea s ai ly cons an . - The numbe o in o ma ion uni s wi h which a pe son can wo k simul aneously, ends o a11 wi hin he ange X=4 oX=7. These indings a e la gely due o he Canadian psycho- logis Pascual-Leone (1972). The immedia e consequence o he o egoing poin s mus be ha , i a lea ning ask makes excessi e demands on a s uden 's wo king memo y, lea ning di icul ies will ensue. We may pos ula e he ollowing: I he numbe o in o ma ion uni s equi ed o a lea ning (o p oblem-sol ing) ask is in excess o he lea ne 's wo king memo y capaci y, he ask canno be sol ed. Johns one and co-wo ke s, in a se ies o ecen in es i- ga ions, ha e p o ided conside able e idence in suppo o his pos ula e. In one o hei in es iga ions (Johns- one and El-Banna 1986), hey a emp ed o es ablish a ela ion be ween i. he wo king memo y capaci y o s uden s, ii. he heo e ical complexi y o a ange o chemical p oblems, measu ed in e ms o he in o ma ion uni s equi ed o hei solu ion, and iii. he p opo ion o s uden s who we e success ul in sol ing hese p oblems. The esul o he s udy which was ca ied ou on 471 uppe seconda y and uni e si y s uden s, a e summa - ized in igu e 4. I is seen ha he success a e in he p oblem-sol ing si ua ions s ongly dec eases as soon as he numbe o he in o ma ion uni s in ol ed in he p oblems exceeds he wo king memo y capaci y o he s uden s. ENSENANZA DE LAS CIENCIAS, 1991,9 (2) INVESTIGACI~N Y EXPERIENCIAS DIDÁCTIC Figu e 4 Ue (ili oh ained h Inhn one ;ind El-R;i i i;i ( 1986) I I I I 3 4 5 6 7 8 Numbe o hough s eps needed oi soi ing p oblem *STMcap.=7; +.STMcap.=6; .STMcap.=5 Wha is ue o p oblem-sol ing asks nus , o cou se, be equally ue o lea ning asks. Hence, i is e iden ha lea ning di icul ies a e a ibu able no only o de iciencies in s uden s' long- e m memo y s uc u e, bu also o limi a ions in hei sho - e m o wo king memo y capaci y. Whi e (1988) gi es a simple, bu ins uc i e example o his wi h e e ence o he s a emen "Concen a ed sul- phu ic aeid is a powe ul dehyd a ing agen ". He sug- ges s ha a ela i ely inexpe ienced s uden would p o- bably pe cei e his s a emen as comp ising a la ge numbe o in o ma ion elemen s, e.g., (Concen a ed) (sulphu ic) (acid) (is a) (powe ul) (de) (hyd a ing) (agen ). An expe ienced eache , by compa ison, may pe cei e he same s a emen in a a less agmen ed way, e.g., (Concen a ed sulphu ic acid) (is a) (powe ul) (dehyd a ing agen ). Thus, wha o he inexpe ienced lea ne appea s as a conglome a e o se en o eigh sepa a e in o na ion uni s, is seen by he expe as a se o ou such uni s. The p ocess o b inging oge he elemen s o in o ma- ion in o g oups o pa cels is gene ally e e ed o as 'chunking'. I is h ough he p ocess o chunking ha we educe o ou sel es he complexi y o he in o ma ion ha eaches us. As Whi e (1988) ema ks: "Size and he e o e numbe o chunks pe cei ed in a si ua ion is one o he big di e ences be ween he knowledgeable pe son (expe , eache , adul ) and he igno an (beginne , s uden , child). Almos pa adoxically, an expe inhabi s a simple wo ld han a beginne because he expe b eaks i in o a smalle numbe o meaning ul uni s." The las poin in Whi e's s a emen is insu icien ly app ecia ed by us in ela ion o ou eaching and su s uden s' lea ning. We end no o be awa e o he ac ha ou s uden s a e incapable o di iding he in o ma ion wi h which we con on hem in o he kinds o chunks ha we ecognize in i . The esul is an o e load o hei sho - e m (wo king) memo y and a consequen ial ail- u e o leam he ull message. This si ua ion does no simply apply o e bal in o ma- ion, bu is equally possible in he case o g aphical o diag amma ic in o ma ion. Whi e (1988) has ecen ly illus a ed his ap ly in his book en i led Lea ning Science. The key o helping s uden s o e come lea ning di i- cul ies a ising om he limi a ions in hei wo king memo y lies in helping hem o see in o ma ion in la ge and, hence, ewe uni s (o 'chunks'). 1 wan o p opose h ee main me hods o achie ing his: a. Ca e ul and pu pose ul s uc u ing o in o ma ion gi en o s uden s, including he lea ning and p oblem- sol ing asks se , in o in o ma ion uni s ha a e clea ly ecognizable by hem. (This may be e e ed o as a 'p e- chunking' ope a ion.) b. A oidance in he design o ex s, labo a o y ins uc- ions, examina ion ques ions, e c., o supe luous, un- necessa y in o ma ion. (Such in o ma ion con ibu es o 'memo y-o e load' and, hence, dis ac s om he essen ial con en o he communica ion.) c. Delibe a e aining o s uden s in 'chunking' s a - egies, h ough he sys ema ic analysis o in o ma ion p esen ed o hem and ways in which i can be o ganized and a anged. In his way, hey should lea n o pe o m beyond he limi s o hei own sho - e m memo y capaci y. The e a e nume ous ways and oppo uni ies in al1 b an- ches o science eaching whe e hese me hods can be e ec i ely employed, o he bene i o su s uden s. CONCLUSION ~ In his pape , a b oad analysis has been p esen ed o he main ypes o lea ning di icul y encoun e ed by s u- ENSEÑANZA DE LAS CIENCIAS, 1991,9 (2) den s in science educa ion and he causes unde lying hem. The gene al p emise was adop ed ha , by and la ge, such lea ning di icul ies a e he esul o de icien- cies o o misma ch. The de iciencies may occu in he knowledge s uc u e held by s uden s in hei long- em memo y, o in hei in o ma ion-handling capabill y which is due o limi a ions in sho - e m memo y capaci y. Misma ch may occu when eaching s a egies used by eache s a e in con lic wi h s uden s' na u al lea ning s yles o hei p e e ences o ce ain ins uc ional pso- cedu es. The de ailed discussion has cen ed on he na u e o memo y- ela ed lea ning di icul ies. In his, e e ence was made o he use o cogni i e mapping as a means o un olding de iciencies and peculia i ies in s uden s' knowledge s uc u e (long- e m memo y e ec s), as a s a ing poin o emedying hem. Simila ly, a en ion was d awn o limi a ion in s uden s' sho - e m memo y capaci y as a cause o ailu e in lea ning and p oblem- sol ing asks. Sugges ions we e also ou lined abou how hese memo y- ela ed lea ning di icul ies can be e- duced o a oided by app op ia e pedagogical ac ions. I is impo an o science eache s and educa o s o ecognize and espond o hese a ious causes o s u- den s' lea ning di icul ies. Doing his will no only educe hese di icul ies, bu also make ou eaching mo e e ec i e. 1 hope ha he ideas p esen ed he e p o ide an app op ia e s imulus in his di ec ion. NOTE (1) Since "language and communica ion" issues in science logically o m pa o he "long- e m memo y" a ea and, hence, educa ion ul ima ely conce n he meaning acqui ed by s u- would come wi hin he pu iew o he in o ma ion-p ocessing den s h ough he use o wo ds and sen ences, i can be a guzd model adop ed abo e. Howe e , he e is some me i in acco ding ha such meanings a e dependen on s uden s' knowledge i a sepa a e posi ion, because o i s gene al impo ance. s uc u e. On his basis, "language and communica ion" would REFERENCES ADAR, L., 1969. A Theo e ical F amewo k o he S udy o Mo i a ion in Educa ion. The Heb ew Uni e si y, Je usa- lem (in Heb ew). BROADBENT, D.E., 1975. The magic numbe se en a e i een yea s, in Kennedy, A. and Wilkes, A. (eds.), S udies in Long e m Memo y. (Wiley: London). CASSELS, J.T.R. and JOHNSTONE, A.H., 1977. Language in Chemis y. In Resea ch o he Class oom and Beyond - Repo o a Symposium. The Chemical Socie y (Now: Roya1 Socie y o Chemis y), London. CLIS (Child en's Leaming in Science P ojec ), 1987. CLIS 'n he Class oom - App oaches o Teaching. Cen e o S ii- dies in Science and Ma hema ics Educa ion, Uni e si y o Leeds, UK. DRIVER, R. andERICKSON, G., 1983. Theo ies -in- Ac ioii: some heo e ical and empi ical issues in he s udy o s uden s' concep ual amewo ks in science, S udies in Science Edu- ca ion, Vol. 10, pp. 37-60. EGGLESTON, J.F., GALTON, M.J. and JONES, M.E. 1976. P ocesses and P oduc s o Science Teaching. (Macmillan Educa ion: London). GAGNÉ, R.M., 1965. The Condi ions o Lea ning. (Hol , Rineha and Wins on: New Yo k). GARDNER, P., 1972. Wo ds in Science. ASTEP (Aus alian Science Teache Educa ion P ojec ), Monash Uni e si y, Clay on, Aus alia. ,-. HOFSTEIN, A. and KEMPA, R.F., 1985. Mo i a ing s a egies in science educa ion: a emp a an analysis, Eu opean Jou nal o Science Educa ion, Vol. 7, pp. 221-229. JOHNSTONE, A.H. and EL-BANNA, H., 1986. Capaci ies, demands and p ocesses -a p edic i e model o science educa ion. Educa ion i 2 Chemis y, Vol. 23, pp. 80-84. KEMPA, R.F., 1983. Leaming Theo ies and he Teaching o Science: Implica ions o Science Teache T aining. In Tami , P., Ho s ein, A. and Ben-Pe e z, M., P ese ice and Inse ice T aining o Science Teache s. (Balaban In e na- ional Science Se ices: Philadelphia, USA). KEMPA, R.F. and MARTÍN, M.J., 1989. Modelos Mo i acio- nales y P e e encias de los Alumnos po di e en es Mé odos de Enseñanza de las Ciencias, Ense ianza de las Ciencias, núme o ex a (111 Cong eso), Tomo 2, pp. 37-40. KEMPA, R.F. and MARTÍN DÍAZ, M., 1990a. Mo i a ional ai s and p e e ences o di e en ins uc ional modes in science. Pa 1: S uden s' mo i a ional ai s, In e na ional Jou nal o Science Educa ion, Vol. 12, pp. 195-203. KEMPA, R.F. and MARTÍN DÍAZ, M., 1990b. S uden s' mo- i a ional ai s and p e e ences o di e en ins uc ional modes in science educa ion. Pa 2, In e na ion~l Jou ial o Science Educa ion, Vol. 12, pp. 205-216. KEMPA, R.F. and NICHOLLS, C., 1983. P oblem-sol ing abili y and cogni i e shuc u e -an explo a o y in es iga- ion, Eu opean Jou nal o Science Educa ion, Vol. 5, pp. 171-184. NEEDHAM, R. and HILL, P., 1987. Teaching s a egies o de eloping unde s anding in science. Cen e o S udies in Science and Ma hema ics Educa ion, Cni e si y o Leeds, UK. NUSSBACM, J. and NOVICK, S., 1982. Al e na i e F ame- wo ks, Concep ual Con lic and Accommoda ion: Towa ds a P incipled Teaching S a egy. Ins ~¿c ional Science, Vol. 11, pp. 183-200. NOVAK, .^. and GOWIN, D., 1985. Lea ning How o Lea n. (Camb idge Uni e si y P ess: Camb idge, UK). ENSEÑANZA DE LAS CIENCIAS, 1991,9 (2) 127