Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Development of memory for narratives : effects of encoding variability and age White, William B. 1985

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1985_A2 W44.pdf [ 5.67MB ]
Metadata
JSON: 831-1.0096785.json
JSON-LD: 831-1.0096785-ld.json
RDF/XML (Pretty): 831-1.0096785-rdf.xml
RDF/JSON: 831-1.0096785-rdf.json
Turtle: 831-1.0096785-turtle.txt
N-Triples: 831-1.0096785-rdf-ntriples.txt
Original Record: 831-1.0096785-source.json
Full Text
831-1.0096785-fulltext.txt
Citation
831-1.0096785.ris

Full Text

DEVELOPMENT OF MEMORY FOR NARRATIVES: EFFECTS OF ENCODING VARIABILITY AND AGE by WILLIAM B. WHITE B.A., Western Washington U n i v e r s i t y , 1972 M.A., U n i v e r s i t y of B r i t i s h C olumbia, 1978 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF EDUCATION i n the Department of EDUCATIONAL PSYCHOLOGY AND SPECIAL EDUCATION FACULTY OF EDUCATION We accept t h i s t h e s i s as c o n f o r m i n g t o the r e q u i r e d s t a n d a r d UNIVERSITY OF BRITISH COLUMBIA September 8, 1985 ©William B. White, 1985 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of EDUCATIONAL PSYCHOLOGY AND SPECIAL EDUCATION The University of British Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date OCT. 15, 1985  DE-6(3/81) i i ABSTRACT Reca l l of narrat ive content was studied in a sample of 170 ch i ldren ranging from 5 to 11 years of age. Age range was div ided into three equal i n t e r v a l s . The chi ldren within each in te rva l were randomly assigned to four encoding condit ions (symbolic, i c o n i c , enact ive, and symbolic-rehearsal) so that any e f fec ts of in teract ions between age-affected cogni t ive capac i t ies and d i f fe ren t encoding condit ions could be gauged at 30 seconds and one week (after encoding). Between-ages (within condit ion) and between condit ions (within age) comparisons revealed that age increase was genera l ly , though not uniformly, accompanied by s i g n i f i c a n t r e c a l l advantage. Analyses revealed that e f fec ts of d i f fe ren t encoding condit ions were s u f f i c i e n t l y var iab le across the ages that age advantage was diminished when free r e c a l l performances of 5-7 year old chi ldren in enactive and iconic encoding condit ions were compared to free r e c a l l performances of older ch i ldren (9-11 years of age) in symbolic condit ions of encoding. The resu l t s are discussed in r e l a t i o n to theore t i ca l issues and educational questions. TABLE OF CONTENTS CHAPTER PAGE L i s t of tables v L i s t of f i g u r e s . . . . : v i Acknowledgement v i I Introduction and l i t e r a t u r e review 1 Introduction L i te ra tu re Review II Der ivat ion of research questions and hypotheses 15 III Design and methodology 23 Design 23 Age 24 Encoding 24 Descr ipt ive data 26 I n t e l l e c t u a l a b i l i t y 27 S o c i a l status 28 Method .28 Sample 28 Apparatus 30 The narrat ive 31 Procedure 32 Prel iminary preparation 32 Procedural flow chart 33 "Warmup" 33 "Training session" 34 "Habituation period 35 Experimental condit ion and purposes 36 Experimental condit ion and procedures - 38 Pos t - tes t procedures 41 Post -Test -1 4 Post -Test -2 4 Scoring measures 4 Rubin's uni ts of measurement 4 Rumelhart's story categories 4 Memory under interrogat ion (or cued r e c a l l measurement) 4 Scoring procedures 4 Analys is 5 iv CHAPTER PAGE IV Results , 52 Part A 53 Age and treatment e f fec ts 53 Part B 56 Within age-span comparisons between the c o n d i t i o n s . . . 56 Y age-span comparisons (between condit ions) 58 M age-span comparisons (between condit ions) 61 0 age-span comparisons (between condit ions) 65 Part C 69 Other observed d i f f e r n c e s : exploratory post-hoc analys is 69 A b i l i t y (IQ), SES, and gender var iab les 78 A b i l i t y (IQ) 78 Gender and SES 78 R e l i a b i l i t y of the Free Reca l l Measurement 79 V Discussion 80 Evaluation of hypotheses 82 Part A: age and treatment e f fec ts 82 Part B: condit ion v a r i a b i l i t y within age-span groups 83 Youngest (Y) age-span group 83 Middle (M) age span group 83 Oldest (0) age-span group 84 Part C: exploratory ana lys is 84 Summary evaluation 85 Discussion and educational impl icat ions of t h i s invest igat ion 91 Ins t ruc t iona l considerat ions 100 L imi tat ions and caveat 103 Age sample 103 Free r e c a l l technique 104 Rehearsal e f fec ts 104 Di rect ions for further research 106 Conclusion 107 Bibl iography 1 09 Appendix A: Rationale for the narrat ive content 121 Appendix B: r a t i o n a l for d i rec t ions for e l i c i t a t i o n of free r e c a l l 122 Appendix C: Questions asked about the story .124 Appendix D: Word uni ts from s t o r y . . . 125 Appendix E: Categories 127 Appendix F: Figure 2B 128 V LIST OF TABLES TABLE PAGE I Means and standard deviat ions of ages ( in months) for each age-span group 29 II Means and standard deviat ions of a b i l i t y (IQ) scores by age-span group and gender 30 III Means and standard deviat ions of each measure at PT-1 and PT-2 of the Y age-span group in each condit ion 59 IV Summary of the resu l ts of a l l pairwise comparisons on the PT-1 category and PT-2 category and cued r e c a l l measures (Y age-span) 61 V Means and standard deviat ions on each measure at PT-1 and PT-2 of the M age-span group in each condit ion 62 VI Summary of the resu l t s of a l l pairwise comparisons on the PT-1 category measure and the PT-2 category and cued r e c a l l measures (M age span ) 65 VII Means and standard deviat ions of each measure at PT-1 and PT-2 by the 0 age-span group in each condi t ion 66 VIII Summary table of resu l t s of the Tukey HSD post-hoc ana lys is on the category measure (at PT-1) with a l l age-spans across a l l condit ions 75 IX Summary table of the resu l t s of the Tukey HSD post-hoc ana lys is on the category measure (at PT-2) with a l l age-spans across a l l condit ions 76 X Summary table of the resu l t s of the Tukey HSD post-hoc ana lys is on the cued r e c a l l measure (at PT-2) with a l l age-spans across a l l condi t ions 77 v i LIST OF FIGURES FIGURE PAGE 1 Procedural flow chart with time elapsed for each a c t i v i t y in prel iminary preparations 33 2A Mean r e c a l l performance (PT-1 and PT-2) on the category measure by each treatment condit ion group (EC IC SC and SIRC) for each age span group 57 2B Mean r e c a l l performance (PT-2) on the cued r e c a l l measure by each treatment condit ion group (EC IC SC and SIRC) for each age span group 128 3A Condition group performance means for Y age span (category and unit measures at PT-1) 60 3B Condition group performance means for Y age span (category, unit and cued r e c a l l measures at PT-2) 60 4A Condition group performance means for M age span (category and unit measures at PT-1) 63 4B Condition group performance means for M age span (category, u n i t , and cued r e c a l l measures at PT-2) 64 5A Condition group performance means for 0 age span (category and unit measures at PT-1) 67 5B Condition group performance means for 0 age span (category, u n i t , and cued r e c a l l measures at PT-2) 68 6 Performance means for each age-span group in each condit ion on the category measure at PT-1 72 7 Performance means for each age-span group in each condit ion on the category measure at PT-2 .73 8 Performance means for each age-span group in each condit ion on the cued r e c a l l measure at PT-2 74 v i i ACKNOWLEDGEMENTS Great appreciat ion i s not enough to extend to one who has given me h is e f f o r t , wisdom, and encouragement. I wish to thank Dr. LeRoy D. Travis who has been f i r s t and foremost in helping me to complete t h i s pro ject . As w e l l , a sincere thanks to Dr. J . Conry, Dr. S. Foster , Dr. S .S . Lee, Dr. N. Suzuki , and Dr. M. Westwood. Further , I would l i k e to extend my thanks to the s ta f f and ch i ld ren at Heritage and Highglen elementary schools for the i r time and cooperation. L a s t l y , I would l i k e to express my grat i tude to those who have given me the i r quiet understanding, moral support, and pat ience. To Archie , E l l a , and Lynda, I thank you. 1 CHAPTER I INTRODUCTION AND LITERATURE REVIEW INTRODUCTION "Everybody who i s s k i l l e d at anything necessar i ly has a good memory for whatever information that a c t i v i t y demands." (Neisser, 1982, p. 17) Fa i lu re to r e c a l l information from connected discourse may be highly impl icated in a student 's poor performance in school . Various researchers have explored the re la t ionsh ips of r e c a l l performance to age, coding systems, r e c a l l . s t ra teg ies , c l u s t e r i n g or chunking, mnemonic or mediation techniques, and organizat ion of mater ial into superordinate or subordinate categories (for example, J a b l o n s k i , 1974; Kreutzer , Leonard, & F l a v e l l , 1975; Moely, 1977; Kobasigawa, 1977). In recent years researchers have d i rected at tent ion to young c h i l d r e n ' s (ages 4-7 years) r e c a l l of nar rat ive mater ia l in the form of sentences or br ief s to r ies (Brown, 1975; Glenn, 1978; Horton & M i l l s , 1984; Par is & Lindauer 1976; Stein & Glenn, 1975). This at tent ion to memory for prose i s actua l l y a r e v i t a l i z a t i o n of a t r a d i t i o n which was given substance by the work of B a r t l e t t (1932). B a r t l e t t departed from what became the mainstream s ty le of memory study in that he preferred to study 2 memory for meaningful content rather than memory for nonsense s y l l a b l e s or other d iscrete uni ts which do not together comprise meaningful content. Researchers in t h i s l a t t e r t r a d i t i o n which Ebbinghaus (1964) establ ished have sought r e g u l a r i t i e s in memory by c o n t r o l l i n g between subject 's d i f ferences in such factors as pr io r learning (D iS ib io , 1982). By subst i tu t ing unconnected uni ts such as nonsense s y l l a b l e s or numerals for the content of connected d iscourse, greater experimental contro l was exercised and greater prec is ion was gained in gauging such matters as memory capacity for d isc rete items; e f fects of organizat ional mnemonics; d i f fe ren t interference phenomena; and rehearsal e f fec ts (to l i s t but a few of the matters s tudied) . In contrast , B a r t l e t t was prepared to accept a lesser degree of prec is ion in order to gain an understanding of memory for meaningful content. The B a r t l e t t t r a d i t i o n has been r e v i t a l i z e d recently as people l i k e Neisser (1982) have concluded that r e g u l a r i t i e s revealed by rigorous studies of the Ebbinghaus t r a d i t i o n t e l l us l i t t l e about the concerns that draw us to the study of memory in the f i r s t p lace . Neisser , l i k e B a r t l e t t , contends that d i g i t span andv word l i s t memory t e l l us l i t t l e about memory for the content of connected discourse upon which teachers re ly for i n s t r u c t i o n . The study of memory for narrat ive content i s addressed here. This i s important for both theore t i ca l and educational purposes because l i t t l e i s known about how such memory develops 3 or how such memory of people, in varying degrees of maturity , i s af fected by d i f f e r e n t encoding cond i t ions . The reported research on memory for narrat ive content evoked suggestions for the present work which compared how memory for narrat ive content var ied with age and encoding cond i t ions . Accordingly, a considerat ion of published research pert inent to the present study i s in order. LITERATURE REVIEW The present study of macro-memory i s an expression of a growing i n t e r e s t , within the research community, in a var ie ty of aspects of macro-memory. Some researchers study memory for tex t . Drum (1985) i s one of these. She of course fol lows in the t r a d i t i o n of B a r t l e t t (mentioned above) and those l i k e Cofer (1941) who kept t h i s l i ne of thought a l i v e . Cofer (1941) was interested in d i f ferences in verbatim learning as compared to l o g i c a l or essent ia l idea learning by col lege students. He found that verbatim r e c a l l of d i f f e r e n t lengths of prose passages was more d i f f i c u l t than was r e c a l l of g i s t or essent ia l ideas. Of course Jerome Bruner and h is col legues have maintained an interest in memory for meaningful content too. Bruner and Olson (1977-78) for example, explored "symbols and texts as too ls of i n t e l l e c t " , because of an abiding interest in memory (and for other considerations connected with " t ransfer" and competence of higher order processes) . 4 Others have focused on s o - c a l l e d schemata theory as i t perta ins to r e c a l l of meaningful content. Schemata of s t o r i e s , or c l a s s i f i c a t o r y systems of structures and elements (often ordered in h ierarchies) (Thorndyke, 1977), have been used to study what i s remembered in a s tory . Among those who report such work are Mandler and Johnson (1977); Rumelhart (1975); Thorndyke, (1977); and Thorndyke and Hayes-Roth (1979), to l i s t but a few. These researchers general ly report that "high l e v e l " (or cent ra l ) story elements are reca l led more than are "low l e v e l " (or per ipheral ) d e t a i l s . Unfortunately, t h i s l i n e of work has come under a cloud in recent times (Horton & M i l l s , 1984) . •Memory f o r . s c r i p t s has also become a topic of in terest to a number of researchers such as Mandler and "Murphy (1983); Bower, Black, and Turner (1979). While " s c r i p t " theory i s a t t r i b u t e d to s p e c i f i c elaborat ion of the frame theory of Minsky (Bower, Black, & Turner, 1979), a " s c r i p t " i s described as a generic memory structure of stereotyped s i tuat ions (such as going to a restaurant ; going to a zoo; r i d i n g a plane, and the l i k e ) (Bower, B lack, & Turner, 1979). Although there may be some question about the use of " s c r i p t s " for study in r e c a l l of prose (Mandler and Murphy, 1983), i t i s an area that some researchers f ind promising as a means of enhancing our understanding of what i s r e c a l l e d . For example, Mist ry and Lange (1985) recent ly reported that s to r ies which contain "strong" s c r i p t (mater ial which t r e a t s experience and events in stereotype) were r e c a l l e d 5 better than s to r ies with "weak" s c r i p t s . That i s , since ch i ldren usual ly v i s i t fast food out le ts more frequently than they v i s i t c i rcuses or zoos they are more l i k e l y to remember a story about a v i s i t to McDonald's than about a v i s i t to the zoo. Attent ion to what i s remembered from complex, meaningful mater ia l has also a t t rac ted i n t e r e s t . Of course g is t continues to be of great i n t e r e s t . B a r t l e t t (1932) reported that the s to r ies he used were not reca l led verbatim but were reca l led in what he and aformentioned researchers would now say was s c r i p t , schemata, or g i s t . G i s t , of course, i s not verbatim content, but i s what Cofer (1941) c a l l e d l o g i c a l and essent ia l ideas of the prose r e c a l l e d . Many researchers have studied sentence, story , and text r e c a l l , and have proposed various concepts of what i s r e c a l l e d and why. Hertel (1985) i s one of those who brings in terest in interference to the study of complex prose m a t e r i a l . This i s merely one of a var ie ty of t h e o r e t i c a l perspectives one n o t i c e s . Thus one sees a range from a Gestal t view of sentence memory to an assocc iat ive view (or from we l l formed wholes to independently l inked concepts) (Horton & M i l l s , 1984, p. 385) of s c r i p t , schemata, and g i s t adapted to endeavors which look for reasons why s p e c i f i c information in sentences and s to r ies i s r e c a l l e d . A l l of t h i s has made the Ebbinghaus t r a d i t i o n look quaint and a rcha ic . The present thread of research, which combines in terest in 6 memory for story content with concerns about the s ign i f i cance or impact of encoding condit ions and the impact of development, i s part of t h i s renewed and growing interest in macro-memory. Some of the roots which anchor, and the main features of the stem which supports the present work warrant mention. Par is & Lindauer (1977) describe studies of r e c a l l of narrat ives in which young ch i ld ren (around 5 years of age), in comparison with older ch i ldren (around 10 years of age), exh ib i t poor r e c a l l of narrat ive m a t e r i a l . This i s consistent with other memory research (eg. on paired assoc iates , d i g i t span, e tc . ) wherein r e c a l l i s general ly reported to increase with age (Brown, 1975a; 1975b; E l k i n d , 1971; F l a v e l l , 1971; J a b l o n s k i , 1974; Kobasigawa, 1977).. However, some research suggests that encoding condit ions also a f fec t r e c a l l (Paris & Lindauer, 1976; 1977; Travis & White, 1979). This suggests that the magnititude of r e c a l l d i f ferences between older and younger chi ldren may be composed of v a r i a b i l i t y due to the interact ion of developmentally d i f f e r i n g s t r u c t u r a l c h a r a c t e r i s t i c s (Jablonski , ' 1974; Par is & Upton, 1976; see below) and encoding condi t ions . Both the Genevan theor i z ing (Inhelder & Piaget , 1964; Fur th , 1969) and the aforementioned evidence (Paris & Lindauer, 1976; 1977; Travis & White, 1979) suggest that when ch i ldren of about 5 to 7 years of age organize information through overt act ions they increase the i r capacity to r e c a l l i t . Apparently, the younger c h i l d ' s cogni t ive funct ioning i s more dependent upon enactive schemes than i s that of older ch i ldren who general ly 7 seem to be comparatively more adro i t in a s s i m i l a t i n g information to f i g u r a t i v e aspects of both imaginal and symbolic representation schemas (Furth, 1969). Both Piaget (Piaget, 1962; 1967; 1969; 1976a; 1976b; Piaget & Inhelder, 1971; 1973) and Bruner (1964; 1966; 1973) have claimed that representat ional systems develop in a sequence: F i r s t , the c h i l d organizes and represents information through h is ac t ions ; the second, or what Bruner (1964; 1973) c a l l s the iconic system, develops from the funct ioning of the enactive or f i r s t system, and i t e n t a i l s the a s s i m i l a t i o n of information to images with which the c h i l d represents information to himself . Th i rd , and s t i l l l a t e r in ontogeny, ch i ld ren develop the capacity to ass imi la te information to more abstract representations the elements of which are signs or true symbols (Inhelder & Piaget , 1964). Each of these representat ional systems putat ive ly enables the c h i l d to construct knowledge of the world; but the knowledge which i s so constructed i s given character by the pa r t i cu la r system or systems which are used to construct i t . While young c h i l d r e n ' s knowledge of the world i s constructed f i r s t , from the i r overt act ions in and on i t (and i t on them), older ch i ld ren can construct a world of images and s t i l l older ch i ld ren can construct knowledge with symbolic representations as wel l as act ions and images (Bruner, 1973; P iaget , 1976). A c o n s t r u c t i v i s t theory (D iS ib io , 1982) such as that of 8 Piaget (Par is & Lindauer, 1977), would suggest then, that not only do ch i ld ren who can construct knowledge with images and symbols gain advantage through increased capac i t ies to know more of the world, they can ass imi la te i t and accommodate to i t with more v e r s a t i l i t y (Piaget, 1976). The capacity to construct knowledge in imagery and symbols apparently enables ch i ld ren to gain increasing independence from the i r concrete act ions for the i r construct ion of knowledge (see a lso D i S i b i o , 1982). From a c o n s t r u c t i v i s t vantage po in t , then, one can argue that what a c h i l d knows is a f fected by both how he or she can know and how he or she has come to know. Moreover, the c h i l d ' s r e c a l l and knowledge are not independent of one another. Accordingly , i f the young (pr ior to age 7 or 8) c h i l d ' s knowledge i s predominantly enactive knowing, one might expect that such ch i ld ren who construct the i r knowledge of given mater ia l through the i r own motoric act ions on i t , w i l l r e c a l l more of the mater ia l than w i l l t h e i r cohorts who are deprived of the opportunity to encode the same mater ia l in the same manner. Older ch i ld ren who are deprived of the opportunity to encode the same mater ia l in enactive schemes, should not be handicapped. With the capac i t ies to construct knowledge of the mater ia l in iconic and/or symbolic schemas more f u l l y developed, ch i ldren at or beyond the age of mid- latency should be able to construct and r e c a l l knowledge of the mater ia l as we l l as or better than the younger ch i ld ren who constructed enactive knowledge of the m a t e r i a l . This l i n e of reasoning seems to inform the Brunerian 9 conjecture (1964) about the ontogenetic order in the development of the representat ional systems. In passing, we should notice that a l l systems are usual ly in place in rudimentary forms by the end of the second year. However, they do not seem to d i f f e r in the i r subsequent careers from other systems, such as s k e l e t a l and neurological systems, which are incomplete in the infancy years. That i s , they w i l l become more elaborate and complete through the:passage of t ime. Apparently, our experience contr ibutes something to such d i f f e r e n t i a t i o n , organizat ion and elaborat ion of language just as i t contr ibutes to the development of coordinated act ions and neurological development (Bower, 1979). However, some systems "have an inherent potent ia l to develop a much greater complexity and capacity to handle a much greater d i v e r s i t y of experience than do other systems. The range of phys ica l act ions which can emerge, be organized and become coordinated seems to be more l i m i t e d than i s the range of images (because more than act ions can be imagined), and s t i l l more l i m i t e d than i s the range for abst rac t ions . For these reasons we can expect the action-scheme systems to approach the i r f u l l p o s s i b i l i t i e s sooner than the image and symbol systems. That i s the i r premier status and l i m i t e d range forcast t h i s . The imaginal systems might be expected to approach the i r f u l l maturity af ter the act ion systems but before the symbol systems. The symbolic capac i t ies have the greatest po tent ia l for e laborat ion , d i f f e r e n t i a t i o n , and organizat ion ; are the most f l e x i b l e ; and, in addit ion to the 10 factors mentioned, because the i r rudiments appear l a s t , they can be expected to take the greatest amount of time and experience to reach f u l l bloom. If Bruner i s co r rec t , one might expect that ch i ld ren in la te latency might be expected to r e c a l l information reconstructed 'from enactive or i c o n i c representation systems with p ro f i c iency . However, the i r capaci ty to reconstruct a narrat ive in a r e c a l l task would be expected to be greater , i f at t h i s age, they construct the i r knowledge of the narrat ive in a symbolic representation system such as language. For as they approach puberty, c h i l d r e n ' s p ro f i c iency with language has usual ly developed s u f f i c i e n t l y to enable them to contemplate p o s s i b i l i t i e s , i d e a l s , and counte r - fac tua l information. This enables them to be more f l e x i b l e , independent, and f l u e n t . Reca l l of language-construed content (as opposed to content ordered with act ions or images) might be expected to be optimal since "mater ial that i s organized in terms of a person's own in terests and cognit ive structure i s mater ia l that has the best chance of being accessib le in memory" (Bruner, 1973, p. 412). Well before t h i s .stage of development the c h i l d ' s capacity to order r e a l i t y with what Pavlov, as w e l l as Vygotsky (1978), c a l l e d the second s ignal system has been repeatedly shown to be wel l es tab l i shed . However, with each passing year our verbal fluency and f l e x i b i l i t y increases very not i ceab ly . Accordingly , by age 10 or so our vocabularies and our knowledge surpass those of pre-school and primary school c h i l d r e n so that we can handle 1 1 feats of symbolic manipulation that myst i fy ch i ldren of 5 or 6 years . What Bruner ( 1 9 6 6 , pp. 4 4 - 5 0 ) c a l l e d the "power" and "economy" of the symbolic "mode" of representation enable one to hold and process more information in that mode than i s the case with the other two (enactive and iconic) representat ional systems. Some research indicates that the development of competence in motoric representation precedes the development of competence with iconic and symbolic representat ions ; that iconic knowing develops p r io r to the development of our, powers of symbolic knowing, and that our memory performances are influenced by our competencies for knowing (Bruner, 1 9 6 4 ; 1 9 6 6 ; 1 9 7 3 ; F l a v e l l , 1 9 6 3 ; Furth , 1 9 6 9 ; Inhelder & P iaget , 1 9 6 4 ; Richardson, 1 9 6 9 ; Vygotsky, 1 9 7 8 ) . L ikewise, t h e r e — a l s o are ind icat ions of notable gains in cogni t ive funct ioning as manifested on a var ie ty of ind icators during the per iod between ages 7 and 8 (Cavanaugh & Borkowski, 1 9 8 0 ; Dunham Sc Lev in , 1 9 7 9 ; F l a v e l l , 1 9 7 1 ; Hartup Sc Coates, 1 9 7 2 ; Inhelder S< Piaget , 1 9 6 4 ; 1 9 6 9 ; J a b l o n s k i , 1 9 7 4 ; Lange Sc Jackson, 1 9 7 4 ) . At th i s time c h i l d r e n ' s s t ra teg ic th ink ing (with mnemonics, organizat ion , •cueing, rehearsa l , e tc . ) and t h e i r ove ra l l a b i l i t y to r e c a l l information improves not iceably ( F l a v e l l , 1 9 7 7 ; Jab lonsk i , 1 9 7 4 ) , as do the i r l o g i c a l competencies which require the maturation of capac i t ies to use true symbols to understand and represent more complex order (Owen, Froman, Sc Moscow, 1 9 8 1 ; P iaget , 1 9 7 6 ) . 1 2 Thus i f young c h i l d r e n ' s iconic and symbolic representat ional structures are r e l a t i v e l y undeveloped, condit ions which require that they encode information in iconic or symbolic schemas may produce a s s i m i l a t i o n d i f f i c u l t i e s which are r e f l e c t e d in the i r comparatively poor r e c a l l of m a t e r i a l . In most inves t igat ions , the subjects ' l a t i t u d e for encoding the mater ia l on which the i r r e c a l l i s to be tested appears to be r e s t r i c t e d by the condit ions in which they t y p i c a l l y encounter the information. Since the information usual ly i s apprehended as i t i s made a v a i l a b l e , in v i s u a l or verbal form, the apparent r e c a l l advantage that accompanies increasing age may simply r e f l e c t concomitant increases in encoding capac i t ies which accompany the development of iconic and. symbolic representat ional systems. If the condit ions in which the mater ia l i s encountered r e s t r i c t encoding to e i ther iconic or symbolic reg is te rs ( i . e . to those systems which are more f u l l y developed with age), the resu l t s may be misleading with regard to the conclusion regarding the memory capacity of ch i ldren at younger ages. At younger ages, c h i l d r e n ' s optimal r e c a l l e f f i c i e n c y may depend upon enactive encoding which usual ly appears to be excluded by the condit ions under which they encounter the mater ial t o - b e - r e c a l l e d in conventional test cond i t ions . Reported r e s u l t s may not represent the young c h i l d r e n ' s r e c a l l capac i t ies w e l l . Measures of free r e c a l l of narrat ive mater ia l require that the young c h i l d reconstruct the narrat ive mater ia l in a symbolic 13 system ( i . e . language). From a c o n s t r u c t i v i s t perspective the young c h i l d ' s capacity to reconstruct the narrat ive in the r e c a l l task might be expected to be greater i f the c h i l d constructs knowledge of i t in h is or her enactive representation system, rather than i f the c h i l d constructs h is or her knowledge of the narrat ive in the symbolic system (see Brown, 1975). This expectation i s based on the premise about the re la t ionsh ip between what one knows and what one can evoke from memory. Given the r e l a t i v e l y immature condit ion of the young c h i l d ' s capacity to construct knowledge in symbolic systems (compared with h is capacity to construct knowledge enact ively ) the mater ia l t o - b e - r e c a l l e d w i l l be known less wel l - i t . w i l l be ass imi lated less f u l l y - i f such knowledge must be constructed in a symbolic reg is ter than i f i t were ass imi lated to enactive schemes. In the f i r s t case, the c h i l d might be induced to say more than he knows; and in the second case, he w i l l know more than he can say - even though he may be expected to r e c a l l more than he would in the f i r s t case. Presumably, one can, at any age, make better t rans la t ions or t ransposi t ions from one reg is ter to another i f one knows wel l the to -be - t rans la ted or transposed ( i . e . transformed and reconstructed) mater ia l wel l than i f one knows the mater ia l less w e l l . If optimal r e c a l l of information by younger ch i ldren depends upon enactive encoding of i t , optimal r e c a l l of such information may be obtained by providing young ch i ld ren with condit ions which enable them to construct knowledge of 14 t o - b e - r e c a l l e d mater ia l through motoric organizat ion of i t . Moreover, since older ch i ld ren become more adept with an iconic representat ional system as Bruner (1964) conjectured, the i r optimal r e c a l l may be associated with condit ions which enable them to ass imi la te t o - b e - r e c a l l e d mater ia l to imaginal s t r u c t u r e s . For s t i l l older c h i l d r e n , condit ions which enable them to benefi t from the advantages of symbolic coding may f a c i l i t a t e optimal r e c a l l . In other words, optimal r e c a l l may depend at f i r s t on enactive encoding; next i t may depend upon i con ic coding; and s t i l l l a t e r , i t may depend upon symbolic coding. What has not been examined extensively i s the degree to which c h i l d r e n ' s memory of narrat ive mater ia l i s af fected by the condi t ions under which they encode i t and how the i r r e l a t i v e e f f i c i e n c y with and d i spos i t i ons to employ d i f f e r e n t representat ional systems change as they develop. Moreover, neither the age-span with in which optimal r e c a l l i s associated with enactive organizat ion or when optimal r e c a l l becomes associated with iconic or symbolic encoding has been e s t a b l i s h e d . In sp i te of the wealth of research on c h i l d r e n ' s r e c a l l a b i l i t i e s , there i s a poverty of information about how or i f age and encoding condit ions in teract to inf luence free r e c a l l . 1 5 CHAPTER II DERIVATION OF RESEARCH QUESTIONS AND HYPOTHESES The foregoing d iscussion suggests that optimal r e c a l l may be associated with d i f f e r e n t encoding condit ions as age increases. Developmental theor is ts such as Piaget and Bruner propose a developmental sequence in the emergence of increasingly mature powers of representat ion. These theor i s ts c laim that while optimal cognit ive funct ioning depends at f i r s t on enactive schemes i t subsequently depends upon information structures and functions associated with images and then symbols. Although the rudiments of a l l three representat ional systems for most ch i ldren may wel l be funct ional by the middle of the second year; and although a l l three systems may be funct ional thenceforth, development of d i f f e r e n t a t i o n in abstract systems i s putat i ve ly less a matter of prospect as age increases (Bower, 1979; Bruner, 1964). That i s , in the preschool years, the development of the ontogenet ical ly premier (enactive) system is more advanced than i s that of the second ( iconic) system and the second i s more f u l l y developed than the t h i r d (symbolic) . Accordingly , the necessity of re ly ing on enactive knowing decreases with increasing age because a l te rnat i ves become both increasingly ava i lab le and r e l i a b l e as 1 6 the second and t h i r d systems develop more f u l l y (with increasing age). This l i n e of t h e o r e t i c a l speculation (eg. Bruner, 1966; 1973) and accumulated evidence (Travis & White, 1979) led to the formulation of the fo l lowing quest ions: (1) Does the r e c a l l advantage that age advance seems to confer remain constant within as wel l as across various encoding condit ions? (2) Does the comparative r e c a l l advantage which enactive encoding condit ions seem to confer on young ch i ldren also emerge for older ch i ld ren when the performances of such older c h i l d r e n , in various encoding condi t ions , are compared? The superior memory performances of young ch i ld ren induced to encode and organize information through overt act ions which represent the tb-be-remembered subject matter (Travis & White, 1979) further suggest that r e c a l l i s af fected by encoding condit ions and the i r in te rac t ion with s t r u c t u r a l var iab les (or cognit ive functions) which apparently change with increased age. According to Piaget ( F l a v e l l , 1963) and Bruner (1964), enactive representations appear to e n t a i l organizat ion of information into systems b u i l t on act ion schemes, and young c h i l d r e n ' s symbolic representations are presumed to develop from the i r p r io r construct ion and use of enactive representations (Bruner, 1964; Inhelder & Piaget , 1964). Unfortunately , there i s presently no basis for assuming that the pattern of r e l a t i v e impact of the encoding condit ions (which were compared in the Travis & White study) i s s i m i l a r or uniform across ages. 17 However, one can conjecture that the improved memory performances of chi ldren in the l a t t e r years of primary school may be a consequence of enhanced maturity of capac i t ies for representing information in iconic or symbolic codes ( Jab lonski , 1974; K a i l , 1979; Kuhlman & Wolking, 1972; Vygotsky, 1978). Notable observers have described how ch i ld ren appear to develop increased capac i t ies to ass imi la te information to iconic and symbolic systems and increase the i r strategy e f f i c i e n c y in learning with increased age ( F l a v e l l , 1970; 1977; Par is & Lindauer, 1976; 1977). In other words, ch i ldren in the l a t t e r years of primary school may have matured or developed representat ional systems which are more economical and powerful than that which i s establ ished through the motoric organizat ion of the i r experiences (see Inhelder, Bovet, & S i n c l a i r , 1974; Inhelder & Piaget , 1964; Lev in , Ghatala, DeRose, Wi lder , & Norton, 1975; Piaget , 1969; Richardson, 1969; Vygotsky, 1978). Thus, older pupi ls may not need to re ly so heavi ly on enactive schemes for the construct ion of knowledge as do the i r younger counterparts. The foregoing analys is led to the formulation of research hypotheses. The tes t ing of these enta i led the organization of equal in te rva l age-spans which together ( in sum) encompassed (1) a period when the u t i l i z a t i o n of enactive schemes seems to be associated with maximum r e c a l l ; (2) a period when the u t i l i z a t i o n of symbolic systems could be expected to be 18 associated with maximum r e c a l l ; and (3) a period which fol lows the former and precedes the l a t t e r , and in which notable t r a n s i t i o n s in cognit ive funct ioning have been documented. The experimental procedures also required that , in addi t ion to design features which systemat ical ly vary the (age-related) s t ruc tu ra l factors (represented by age as the i r proxy) , factors which inf luence encoding also had to be var ied and standardized. Accordingly , four experimental condit ions were devised to induce encoding v a r i a b i l i t y within each age-span sub-set of a sample of c h i l d r e n . These features, together with random assignment within age-spans; standard (common) information (narrat ive) content; standard r e c a l l metr ics ; and uniform time cont ro l s , comprised the major features of the design devised to test the hypotheses by experiment and gain data which bear on the research questions. Therefore, given three age-spans (Y= 60-83 months; M= 84-107 months; 0= 108-131 months); given four encoding treatment condit ions designed to allow for est imation of the impact of such condit ions both with in and between age-spans; and given a common narrat ive as wel l as common measurement instrumentation, one can test predict ions with respect to treatment (condition) e f fec ts within and between age-spans (and age e f fec ts within treatments). Thus, the fo l lowing hypotheses (H) were formulated with regard to the measured r e c a l l performances of the twelve sub-sets of the sample: An in teract ion of age and encoding condit ions i s suggested 19 by two propos i t ions : F i r s t , as age increases c h i l d r e n ' s r e c a l l of narrat ion content i s more complete when such content i s organized and represented with symbols (such as language) than i s the case when such content i s organized and represented with enactive and iconic representat ions. Second, at younger ages, ch i ldren are better able to r e c a l l mater ia l that i s organized and represented by enactive schemes (compared to r e c a l l of mater ia l organized and represented by icon ic or symbolic s t ruc tu res ) . Thus the f i r s t hypothesis was formulated as fo l lows : H1: There w i l l be a s i g n i f i c a n t in te ract ion between age-spans and encoding condit ions on r e c a l l . Since there i s l i t t l e reason or evidence which suggests that enactive c a p a b i l i t i e s which have been associated with optimal r e c a l l by young ch i ldren should decl ine with age; and since r e c a l l d i f ferences were expected to be observed (as past research indicates improvements of r e c a l l with age increase) , the fo l lowing hypotheses were made with regard to age e f f e c t s : H2: The oldest (0) age-span w i l l be s i g n i f i c a n t l y superior to the youngest and middle (Y and M) age-spans on r e c a l l . H3: The middle (M) age-span w i l l be s i g n i f i c a n t l y superior to the youngest (Y) age-span on r e c a l l . S ince, as indicated above, the enactive c a p a b i l i t i e s (for r e c a l l performance) of the ch i ld ren are not expected to decl ine with age, but the symbolic c a p a b i l i t i e s (for r e c a l l performance) are expected to increase with age, the fo l lowing hypotheses were 20 made for condit ion e f f e c t s : H4: The r e c a l l performances of ch i ldren from enactive encoding condit ions w i l l be superior to the r e c a l l of ch i ldren from iconic encoding condi t ions . H5: The r e c a l l performances of ch i ldren from enactive encoding condit ions w i l l be superior to the r e c a l l of the ch i ld ren from symbolic encoding condi t ions . The next set of hypotheses was concerned with what educators must deal with when considering the p o s s i b i l i t y of var iab le e f fec ts of learning condit ions on the performances of ch i ld ren between the d i f fe ren t encoding condit ions at s p e c i f i c age l e v e l s . For example, teachers of ch i ld ren 9 to 11 years of age, while perhaps in te res ted , presumably would not be as concerned with the performances of ch i ldren 5 to 7 years of age as they would be with performances in the 9 to 11 year age-span. However, l i t t l e research has been conducted on c h i l d r e n ' s performance with regard to encoding condi t ions . Therefore, fo l lowing the t h e o r e t i c a l speculation and accumulated evidence presented above, hypotheses were formulated which are concerned with the d i f f e r e n t i a l attenuation of condit ion e f fec ts by developmental (age) advance. The f i r s t of these fol lows from the premise that observed re la t ionsh ips reported by Travis & White (1979) could be be r e p l i c a t e d . Accordingly , the fo l lowing hypothesis focused on encoding treatment e f fec ts within the youngest age-span: H6: In the Y age-span, ch i ldren from the enactive encoding 21 condit ion w i l l r e c a l l s i g n i f i c a n t l y more narrat ive content than the i r age cohorts in the iconic or symbolic encoding condi t ions . Since there was no empir ica l basis for pred ic t ing that the condit ions would d i f f e r e n t i a l l y a f fec t r e c a l l in the M age-span; and there were theore t i ca l reasons for supposing that the d i f f e r e n t i a t i o n and maturation of cognit ive or symbolic a b i l i t i e s would enable ch i ldren of th i s age to construct representations and r e c a l l same with greater prof ic iency than was the case when they were younger, an exploratory analys is was conducted to test for any systematic d i f ference between the condit ions for t h i s age-span. While youngsters are usual ly remarkably able to handle true symbols in speech at an ear ly age, by age 9 the i r symbolic powers are much r i c h e r . They order abstract ions with f luency, f l e x i b i l i t y and complexity that are beyond the powers of young c h i l d r e n . The economy and power of such symbol systems should confer cognit ive and r e c a l l advantages on those members of th i s age-span, who, due to experimental cond i t ions , re ly on same (as opposed to cohorts who do not do so) . In a s i m i l a r ve in , by th i s age the superior e f f i c i e n c y of iconic processing (compared with enactive) should be observable. Moreover, the often observed advantage of rehearsal ( K a i l , 1979) can also be expected to be seen. Therefore, the fo l lowing hypothesis was made with respect to encoding treatment e f fec ts within the oldest group (0 age-span): 22 H7: In the 0 age-span, ch i ldren from the symbolic and rehearsal encoding condit ion w i l l demonstrate superior r e c a l l performance when compared to the performances of the i r age cohorts from the other (enact ive, i con ic , and symbolic-no rehearsal) encoding condi t ions . Since r e l i a b l e knowledge of the degree of inf luence which d i f f e r e n t encoding condit ions and d i f fe ren t l eve l s of maturity have on free r e c a l l was sought, two sets of observations were made at t h i r t y seconds (PT-1) and one week (PT-2) af ter the t o - b e - r e c a l l e d narrat ive had been encountered. The second comparison (PT-2) enables one to assess retention of the sort which i s of i n t e r e s t in i n s t r u c t i o n . It also provides a basis for assessing r e l i a b i l i t y of measurement. Therefore, one week a f te r PT-1 had been concluded the second post test (PT-2) was administered to test for s t a b i l i t y of observed free r e c a l l performance d i f f e r e n c e s . Enduring e f fec ts are most readi ly seen to have pedagogcial impl icat ions since teachers are concerned with long term retent ion of information; and the e f fec ts of age in in te ract ion with condi t ions , such as might be discerned, can be seen to bear on questions about i n s t r u c t i o n a l p rac t i ces . Therefore, the above hypotheses apply to scores on both the immediate t e s t s of r e c a l l ( t h i r t y second delay) , and the more delayed test of r e c a l l (one week delay) ; that i s , the hypothesized r e l a t i o n s h i p s . were expected to be robust across a period of one week. 23 CHAPTER III DESIGN AND METHODOLOGY DESIGN As suggested e a r l i e r , s i g n i f i c a n t improvements of r e c a l l a b i l i t i e s (as wel l as other cognit ive a b i l i t i e s ) are frequently notable around 8 years of age (Richardson, 1969). However, l i t t l e not ice has been taken of the extent to which the magnitude of measured gains i s af fected by encoding condit ions for these ch i ld ren and those, both younger and o lder , with whom they are compared.-With the above considerations in mind, t h i s study was designed to invest igate how four d i f f e r e n t encoding condit ions af fect free r e c a l l performances of 170 ch i ldren who var ied in age from 5 to 11 years of age. Further , i t was designed to discover the extent to which the free r e c a l l performances for each of the twelve age-encoding var ia t ions would be stable over a period of one week. Accordingly , the r e c a l l performances of ch i ld ren in the age-span 84-107 months were compared with those from the adjacent (older and younger) age-spans of equivalent width, so that a l l age-spans were compared to one another. 24 Ac[e The three age-spans included: (1) the youngest (Y) group (age 60-83 months) the members of which have not reached the age within which notable cognit ive gains are commonly seen (and to which reference has just been made); (2) the middle (M) group (age 84-107 months) which included those one might expect to manifest the memory gains which accompany other gains in cogni t ive powers; (3) the oldest (0) ch i ldren (age 108-131 months) who might be expected to be the most mature thinkers and p ro f i c ien t r e c o l l e c t o r s . Encoding Chi ldren within each of three adjacent age-spans were randomly assigned to the four encoding condi t ions , three condit ions of which were devised to r e s t r i c t encoding to one or another of the representat ional systems (enactive, i c o n i c , and symbolic) ; and.one of which (symbolic and imagery rehearsal) was included to make possible the estimation of the magnitude of imagery rehearsal e f f e c t s . These encoding condit ions are denoted by the fo l lowing four names and abbreviat ions: (1) enactive (EC); (2) iconic ( IC); (3) symbolic (SC); (4) symbolic and imagery rehearsal (SIRC). The encoding condit ions were designed to r e s t r i c t encoding opt ions, such that (a) a sub-set of each age group encountered 25 c o n d i t i o n s which were d e s i g n e d t o maximize the p r o b a b i l i t y t h a t the t o - b e - r e c a l l e d i n f o r m a t i o n would be t r a n s f o r m e d i n t o a system of a c t i o n schemes (an e n a c t i v e system of r e p r e s e n t a t i o n ) ; (b) a n o t h e r sub-group was induced t o t r a n s f o r m the i n f o r m a t i o n i n t o systems of mental imagery ( i c o n i c schemata or an i c o n i c system of r e p r e s e n t a t i o n ) ; (c) a n o t h e r sub-set of each age-group was i n duced t o employ the system of m e n t a t i o n we c a l l s y m b o l i c (or a s y m b o l i c system of r e p r e s e n t a t i o n ) ; (d) and the l a s t of the sub-groups f o r each age-span was induced to employ the system of s y m b o l i c r e p r e s e n t a t i o n and then imagery r e h e a r s a l . T h i s f o u r t h c o n d i t i o n was i n c l u d e d because c h i l d r e n i n the e n a c t i v e c o n d i t i o n would n e c e s s a r i l y r e h e a r s e the substance of the s t o r y as they d r a m a t i z e d i t , and c h i l d r e n i n the a l t e r n a t i v e ( i c o n i c and s y m b o l i c ) t r e a t m e n t s d i d not have t o r e h e a r s e the m a t e r i a l . A c c o r d i n g l y , an e s t i m a t e of the e x t e n t t o which r e h e a r s a l i n f l u e n c e s the r e c a l l of those i n the e n a c t i v e c o n d i t i o n was sought. The f o u r t h c o n d i t i o n t h e n , c o n s i s t e d of the c o n d i t i o n s of the s y m b o l i c t r e a t m e n t p l u s a f e a t u r e t h a t a l l o w e d f o r an e s t i m a t i o n of r e h e a r s a l e f f e c t s . A comparison of the r e c a l l performances of c h i l d r e n i n t h i s ( s y m b o l i c and r e h e a r s a l ) , c o n d i t i o n w i t h those i n the s y m b o l i c c o n d i t i o n s h o u l d i n d i c a t e the magnitude of the impact of r e h e a r s a l on r e c a l l . T h i s d i f f e r e n c e , s u b t r a c t e d from the e n a c t i v e c o n d i t i o n performance s c o r e s , s h o u l d a l l o w one t o e s t i m a t e the magnitude of r e h e a r s a l c o n t r i b u t i o n s t o t h e s c o r e s from the e n a c t i v e c o n d i t i o n . A more d e t a i l e d d e s c r i p t i o n of the e n c o d i n g c o n d i t i o n s can be see on pages 36-40 (b e l o w ) . 26 The performances of each sub-set of the sample as measured with two indices of free r e c a l l , were compared with those of a l l other sub -sets . That i s , performance comparisons were made within the age-spans across the condit ions and then within the condit ions across the age-spans. At the end of the second free r e c a l l procedure (during PT-2) , questions were employed to obtain another type of memory measure. This allowed for further assessment of the impact of the d i f f e r e n t encoding condit ions on c h i l d r e n ' s memory for narrat ive discourse as they vary in age. This design afforded the assessment of (1) the extent to which free r e c a l l * var ied with encoding condit ions within age-spans; (2) the extent to which the pattern of r e c a l l performances associated with the condit ions within age-spans remained stable across age-spans; and (3) the extent to which free r e c a l l var ied with age within and across condi t ions . Descr ipt ive Data Although treatment groups within each of the three age-span sub-sets of the sample were formed on the basis of random assignment, random assignment between age-spans (as contrasted to within age-spans) was not possible given the age manipulation requirement of the design. Accordingly , between age-span groups could d i f f e r from one another with regard to not only age but other c h a r a c t e r i s t i c s such as gender composit ion, a b i l i t i e s , and socio-economic c l a s s . Since the r e c a l l performances (as 27 assessed by three measures - see below) may have covaried with gender, a b i l i t i e s ( I .Q . ) , and s o c i a l c lass (SES), descr ipt ive s t a t i s t i c s on these var iab les for each age group provided a basis for assessing the extent to which the groups came from the same gender-mix, socio-economic, and a b i l i t i e s populations. Therefore, as a precaut ion, data on IQ, sex, and s o c i a l c lass were analyzed to c o n t r o l , by s t a t i s t i c a l means, any possible systematic inf luence they might have on the dependent measure scores. A descr ip t ion of the i n t e l l e c t u a l a b i l i t y measure and s o c i a l c lass information fo l lows . I n t e l l e c t u a l A b i l i t y The a b i l i t y measure (I .Q.) was derived by administering Form 2 of the Quick Test (QT) (Ammons & Ammons, 1962). The QT i s a reasonable estimator of the general l eve l of i n t e l l e c t u a l a b i l i t y as indicated by perceptual -verbal performance (Ammons & Ammons, 1962; 1979; Davis & Dizzonne, 1970; Joest ing & Joest ing , 1972; Libb & Coleman, 1971; Mednick, 1969; V i o l a t o , White, & Trav is , 1984), which performance was of interest in t h i s study. Since t h i s study enta i led the p a r t i c i p a t i o n of ch i ld ren 5 to 1 1 years of age, a test that d id not e n t a i l wr i t ten responses ( d i f f i c u l t for many 5 year o ld chi ldren) was required. Moreover, since the design of t h i s study made the use of the same test for a l l pa r t i c ipants des i rab le , and circumstances required that ch i ldren d id not lose more c lass time than was necessary, the at t ract i veness of the QT for the present purposes was enhanced (Dizzonne & Davis , 1973; Gendreau, Roach, & 28 Gendreau, 1973; Nicholson, 1977). Soc ia l Status Parents were asked to indicate the i r usual occupations on a parental permission form. The occupation of the father , or in h is absence, that of the mother, was used to indicate s o c i a l s tatus . After the parents' occupations were es tab l i shed , a l l ch i ldren were c l a s s i f i e d as belonging to one of four SES c l a s s i f i c a t i o n s : 1) Entrepreneurs ( e . g . , independent, se l f employed, businessmen); 2) Professional/Managerial ( e . g . , lawyers, phys ic ians , teachers, executives, e t c . ) ; 3) S k i l l e d Labour ( e . g . , l icensed workers such as plumbers, e l e c t r i c a n s , e t c . ) ; and 4) Non -sk i l l ed (unlicensed) Labour ( e . g . , c l e r k s , labourers, m i l l workers, e t c . ) . Three raters c l a s s i f i e d each parent 's occupation independently; and the i r rat ings were compared. There was 100% agreement between the independent ra t ings . The number and percentage of ch i ldren in each SES l e v e l are presented below in the sample descr ipt ion sec t ion . METHOD Sample The sample of 170 ch i ldren ranged in age from 65 to 131 months. They attended two schools in the same urban school d i s t r i c t in nor th -cent ra l B r i t i s h Columbia; and came from a developed neighborhood which followed c i t y planning bylaws of proport ional (economic) mixed housing development (the mix of 29 d o m i c i l e s one would expect from the o c c u p a t i o n a l mix d e s c r i b e d a b ove). P a r e n t a l consent f o r each c h i l d who was a p r o s p e c t i v e p a r t i c i p a n t was o b t a i n e d . The e n t i r e age range was d i v i d e d i n t o t h r e e e q u a l i n t e r v a l s ( d e s c r i b e d e a r l i e r ) so t h a t t h r e e n o n o v e r l a p p i n g age groups (Y, M, and 0) were formed. Each age group so formed had a minimum of 54 c h i l d r e n . C h i l d r e n from each age group were randomly a s s i g n e d t o one or another of f o u r t r e a t m e n t c o n d i t i o n s . T h i s a l l o w e d f o r a minimum of 13 s u b j e c t s f o r each treatment c o n d i t i o n a t each age i n t e r v a l . The age means (e x p r e s s e d i n months) and s t a n d a r d d e v i a t i o n s f o r each sample age-span can be seen i n T a b l e I . T a b l e I Means and S t a n d a r d D e v i a t i o n s of Ages ( i n months) f o r each Age-Span Group Age-Span Group Mean Age ( i n months) Range Sd n Y 72.66 65-83 4.80 59 M 98.37 85-107 5.55 54 0 120.77 109-131 6.66 57 Total=170 The Quick Test (Ammons & Ammons, 1962) was a d m i n i s t e r e d t o g a i n an e s t i m a t e of each c h i l d ' s i n t e l l e c t u a l a b i l i t y (see Table I I ) . Demographic data were c o l l e c t e d t o e s t a b l i s h a socio-economic r a t i n g f o r each c h i l d ( t h e p e r c e n t a g e and number of c h i l d r e n i n each SES l e v e l were: 5.9% e n t r e p r e n e u r ( n = l 0 ) ; 30 22.9% professional/managerial (n=39); 37.6% s k i l l e d labour (n=64); 33.5% nonski l led labour (n=57)). These data and the gender of each c h i l d were recorded, which provided a basis for i d e n t i f y i n g or descr ibing the populations which th i s sample represents. These descr ip t i ve data resu l t s are reported in Chapter IV. Table II Means and Standard Deviations of A b i l i t y (IQ) Scores by Age-Span Group and Gender Age-Span Gender Mean Score Sd n Group A b i l i t y (IQ) Y M 105.38 10.91 31 . F 101.32 14.41 28 Mean Total 103.46 12.75 59 M M 102.55 10.80 22 F 100.63 10.55 32 Mean Total 101.41 10.59 54 O M 101.82 17.39 28 F 93.79 12.72 29 Mean Total 97.74 15.59 57 Mean Grand Total 100.89 13.32 170 Apparatus A Panasonic portable cassette tape recorder (Model #RQ 2108) with a b u i l t in microphone was used to de l i ver a narrat ive and to record the c h i l d r e n ' s r e c a l l of same. A cassete tape 31 recording of a male voice voca l i z ing a narrat ive for one minute and 17 seconds in durat ion, was used with the recorder. (See below for a t ransc r ip t ion of the narrat ive and Appendix A for the rat iona le for the use of th i s m a t e r i a l ) . Two homemade cotton sock puppets were used. One, a brown cotton sock on which were a f f i x e d f e l t eyes, ears, nose, and t a i l resembled a gopher. It was used during a t r a i n i n g session in which each c h i l d par t i c ipated (see procedural sect ion below). The other was made to resemble a black cougar. It was also made from a (black) cotton sock on which were sewn f e l t eyes, ears, nose, and t a i l . This puppet was used in the EC and IC experimental condi t ions . A two part fo ld ing scene of a forest with a winding t r a i l , a c lus te r of houses for a town, and a backdrop of a blue sky, c louds, and sun was used for a background prop in the EC and IC cond i t ions . The cougar puppet and scenery prop were u t i l i z e d to dramatize the experiences of the cent ra l character in the nar ra t ion . A Remex e lec t ron ic stop watch (Sports Stopwatch model) was used to measure time sequences during t r a i n i n g , treatment, and t e s t i n g . A record form for the Quick Test resu l ts and demographic data, as wel l as a parental consent form, was a lso used for each c h i l d . The Narrat ive The story used in t h i s study was as fo l lows : One ear ly morning, a black cougar named Rufus l e f t h is hole that was home and walked along a forest t r a i l . He was looking for something to eat . As he looked he turned h is head from one 32 side to the other. The longer he looked the faster he walked. He even ran a l i t t l e , with h is head s t i l l turning as he searched. Suddenly the forest disappeared and to h is l e f t he saw a town. He thought, "There must be food there . " He leaped toward the town. Soon he could see nothing but houses when he turned round and round. As he d id so he nearly f e l l over. He f e l t dizzy - and hungry. The sight of a bowl with a ju icy bone in i t made him forget he was dizzy and hungry. "That w i l l make a good meal , " he s a i d . Rufus dived at the bowl. The c l a t t e r awoke a dog whose bark made the cougar's ears stand up. The dog's bark was coming c loser and c l o s e r . Rufus pressed h is body c lose to the ground and began to creep away. But suddenly something made a loud noise next to him. Rufus jumped high into the a i r and ran home as fast as h is legs could carry him. He had nothing to eat for breakfast but food for thought: A bone in a bowl puts a cat in the hole . Procedure Prel iminary preparat ions. Prel iminary preparations consisted of three phases: (1) A one minute "warm up" designed to put each c h i l d at ease; (2) a three minute " t r a i n i n g session" which was devised to ensure that a l l ch i ld ren would (or could) use a puppet to dramatize narrat ive content; (3) a one minute "habituat ion period" wherein the ch i ldren were made fami l ia r with the presence and operation of a tape recorder. A flow chart for times involved on each phase of the prel iminary preparations with each experimental condit ion can be seen in Figure 1 below. 33 Figure 1 Procedural flow chart with time elapsed for each a c t i v i t y in prel iminary preparations EC Preparation Time in Experimental Conditions IC SC SIRC A c t i v i t y "Warm up" "Tra in ing" "Habituat ion' and Treatment Set-up* 1 mm, + 3 min, + 50 sec, + 10 sec 1 min 3 min, 50 sec, + 10 • sec 1 min. 3 min, 1 min. 1 min. + 3 min. + 1 min. Total Time 5 min 5 min. 5 min, 5 min, Treatment set-up for the EC and IC condit ions consisted of p lac ing a scenery prop in front of the c h i l d , as we l l as p lac ing a puppet on the c h i l d ' s hand (EC) or experimenter's hand (IC) - see below for a descr ipt ion of th i s a c t i v i t y . Each of the prel iminary preparation phases i s discussed in tu rn . "Warm-up" Each c h i l d was introduced to the experimenter in the classroom and was then taken to a tes t ing room. As soon as the c h i l d entered the room he/she was asked to s i t down at a t a b l e . 34 A br ie f "warm-up" designed to put the c h i l d at ease ensued ( i . e . the experimenter talked with the c h i l d in a warm and f r iend ly manner about conventional topics which a r i se in i n i t i a l encounters wherein people "get acquainted" with one another) . After one minute lapsed the experimenter s a i d , "I have something I would l i k e to show you." A " t r a i n i n g " session then fol lowed. "Training Session" The t r a i n i n g session lasted three minutes for a l l c h i l d r e n . It enta i led a demonstration of how a puppet could be made to dramatize or make manifest whatever some sentences ind icated . When the demonstration was completed the experimenter asked the c h i l d i f he/she would l i k e to t r y . With a p o s i t i v e response the experimenter gave the c h i l d the puppet and helped put i t on the c h i l d ' s hand i f needed. The experimenter then asked the c h i l d to demonstrate some moves that the puppet might make. After the c h i l d made some manipulations with the puppet, the experimenter asked the c h i l d to dramatize an uttered sentence with the puppet. Two such sentences were used to provide help and observe the c h i l d ' s response. The f i r s t sentence used, was: "The gopher was walking through the grass when he heard h is mother c a l l i n g him to come to dinner as fast as he c o u l d . " The second sentence used, was: "A gopher was walking through a yard when he saw some f r iends . He stopped to say h e l l o , and then went on aga in . " Since the t r a i n i n g session was intended to ensure that the c h i l d was competent to manipulate a puppet while l i s t e n i n g , once the c h i l d responded appropr iate ly , a d iscussion 35 about puppets ensued u n t i l the three minutes lapsed (from the time the t r a i n i n g session began). If there was d i f f i c u l t y in manipulating or dramatizing the sentences, the experimenter helped the c h i l d by either moving the c h i l d ' s hand or making suggestions, and repeating the sentences u n t i l three minutes lapsed from the beginning of the t r a i n i n g sess ion. Any c h i l d who showed a d i s i n c l i n a t i o n to undertake or had d i f f i c u l i t y dramatizing the sentence content at the end of t h i s time frame was omitted from the study. There were four such cases. At the end of the " t r a i n i n g sess ion" , the puppet was removed from the c h i l d ' s hand and placed out of the c h i l d ' s s igh t . A short (approximently one minute) d iscussion about tape recorders ("habituation period") fol lowed. "Habituation Period" The discussion about tape recorders commenced when the experimenter pointed to a tape recorder which had been on the table since the c h i l d entered the room, and s a i d , " ( C h i l d ' s name), th i s i s a tape recorder. Have you ever used one before?" The operation of the tape recorder was then discussed. After about one minute (dependent upon the treatment condit ion the c h i l d had been assigned to - see above flow chart) the c h i l d was t o l d , "I have a story on the tape recorder that I would l i k e you to l i s t e n t o . Would you l i k e to l i s t e n to i t ? " If the c h i l d responded negatively to l i s t e n i n g to the story , he/she was returned to the classroom and subsequently dropped from the study (one Y age group c h i l d d id not wish to l i s t e n to the story 36 and was dropped from th is study at th i s po in t ) . With a pos i t i ve response the session continued. In sum, the introduct ion to the experimenter (one minute in durat ion) , the t r a i n i n g session (three minutes in durat ion) , d iscussion about tape recorders and the i n v i t a t i o n to l i s t e n to a story (about one minute in durat ion) , took a t o t a l of f i ve minutes. This same amount of time was spent with each c h i l d before the story was heard. No c h i l d spent more or less than th i s f i ve minutes with the experimenter on these procedures. Each c h i l d was then given a story in his/her assigned treatment condit ion . Experimental Conditions and Purposes The encoding condit ions (as discussed above and below) were structured to r e s t r i c t encoding options. That i s , each condit ion was designed to enhance the p robab i l i t y that the narrat ive content would be encoded in one system and not in the a l t e r n a t i v e s . These encoding condit ions are denoted by the fo l lowing four names and abbreviat ions: (1) enactive (EC); (2) iconic (IC); (3) symbolic (SC); (4) symbolic and imagery rehearsal (SIRC). The purpose and major feature of each of these fo l low. A more complete descr ipt ion of each condit ion i s provided thereaf te r . The purpose and major feature of each condit ion fo l lows : (1) Chi ldren in the enactive (EC) condit ion encountered condit ions which were designed to maximize the p r o b a b i l i t y that 37 the to-be-recalled information, a narrative, was transformed or transposed into a system of sensory motor schemes (or an enactive system of representation). In this condition children manipulated a puppet to dramatize the content of the narration while they listened to the tape recording of i t . (2) Children in the iconic (IC) condition encountered conditions which were designed to maximize the p r o b a b i l i t y that the to-be-recalled information (a narration) was transformed or transposed into systems of mental imagery (or iconic schemata). In t h i s condition children l i s t e n e d to a narration while watching a puppet (manipulated by an experimenter so that i t enacted the narrative content). (3) Children in the symbolic (SC) condition encountered conditions which were designed to maximize the pr o b a b i l i t y that the to-be-recalled information (a narration) was assimilated to a system of symbolic representations. In t h i s condition children merely listened to the narration. No puppets or scenery props were used. (4) Children in the symbolic encoding and imagery rehearsal (SIRC) condition merely lis t e n e d to the recorded narrative (as did children in SC); but in addition, and immediately after they l i s t e n e d to the story, the SIRC children were told to close t h e i r eyes and t e l l themselves the story they have just heard and try to see what happens in the story as they t o l d i t to themselves (imagery rehearsal). No puppets or scenery props were used. 3 8 Experimental Conditions and Procedures As stated e a r l i e r , a l l c h i l d r e n , within each age-span, were randomly assigned to the four cond i t ions ; and they were treated and tested i n d i v i d u a l l y . Involvement with treatment condit ions commenced when the tape recorder was turned on. Further descr ipt ions of d e t a i l in each condit ion fo l low: ( 1 ) EC: Af ter the c h i l d responded p o s i t i v e l y to the i n v i t a t i o n to l i s t e n to the story , the experimenter placed a scenery prop on the table in front of the c h i l d , brought out another puppet and s a i d , "Here is another puppet. Would you l i k e to put t h i s one on?" When the c h i l d agreed, the experimenter then helped the c h i l d put the puppet on his/her preferred hand when needed. (This procedure took ten seconds, and i s part of the prel iminary preparations described above). The experimenter then s a i d , "I am going to play a story that i s on t h i s tape recorder. I want you to make the puppet act out the story while you are l i s t e n i n g to the s tory . Do you think you can do that?" As soon as the c h i l d gave a pos i t i ve response, the experimenter, while turning the tape recorder on, s a i d , "L is ten very c a r e f u l l y to the s tory , and make the puppet do what the story says." The story was then played. As the story was being played the experimenter watched but d id not make any gestures or comments during th i s t ime. As soon as the story was f i n i s h e d , the scenery was removed from s igh t , and the puppet was taken off of the c h i l d ' s hand and placed out of s igh t . The story tape was removed from the tape recorder; and a blank tape was put in i t s p lace . During t h i s 39 time the experimenter remained qu ie t . The removal of the props and story tape, and the placing of a blank tape into the tape recorder usual ly took s l i g h t l y less than 30 seconds. If th i s procedure took less time, the experimenter remained quiet and waited ( i . e . looked at a procedure paper in front of him, acted busy, and so forth) u n t i l 30 seconds passed. Pos t - tes t 1 (PT-1) then commenced (see below). (2) IC: Af ter the c h i l d responded p o s i t i v e l y to the i n v i t a t i o n to l i s t e n to a story , the experimenter placed a scenery prop on the table in front of the c h i l d , and brought out another puppet used with the story to be heard. The experimenter then placed the puppet on h i s , the experimenter's, hand. (This procedure took ten seconds; and i s part of the prel iminary preparations described above). The experimenter then s a i d , "I am going to play a story that i s on th i s tape recorder. I w i l l make the puppet act out the story while you watch the puppet and l i s t e n to the story . Do you think you can do that?" As soon as the c h i l d gave a pos i t i ve response, the experimenter, while turning on the tape recorder, s tated, "L is ten very c a r e f u l l y to the story and watch the puppet do what the story says . " While the recorded story was played on the tape recorder, the experimenter enacted the story with the puppet._ As soon as the story was over, the same procedures car r ied out when the story f in ished in the EC condi t ion were fol lowed. PT-1 then began (see below). (3) SC: Af ter the c h i l d responded p o s i t i v e l y to the i n v i t a t i o n to l i s t e n to a story , the experimenter s a i d , "I am going to play a story that i s on th i s tape recorder. I want you to l i s t e n to 40 the story while i t i s being played. Do you think you can do that?" As soon as the c h i l d responded p o s i t i v e l y , the experimenter s a i d , while turning the tape recorder on, "L isten very c a r e f u l l y to the s to ry . " While the story was being played, the experimenter l i s tened with the c h i l d . The experimenter followed the same procedures as in the EC and IC condit ions as soon as the story f i n i s h e d , except that there was no puppet or scenery prop to remove. PT-1 commenced when the standard time had elapsed. (4) SIRC: After the c h i l d responded p o s i t i v e l y to the i n v i t a t i o n to l i s t e n to a story , the experimenter s a i d , "I am going to play a story that i s on t h i s tape recorder . I want you to l i s t e n to the story while i t i s being played. Do you think you can do that?" As soon as the c h i l d responded p o s i t i v e l y , the experimenter s a i d , .while turning the tape recorder on, "L is ten very c a r e f u l l y to the s to ry . " While the story was being played, the experimenter l i s tened with the c h i l d . Immediately a f ter hearing the story the ch i ld ren in t h i s group were t o l d , "Close your eyes, t e l l yourself the s to ry , and imagine or t ry to see what happens in the story as you t e l l yourself the s to ry . " The c h i l d was given 30 seconds to do t h i s . While the c h i l d was doing t h i s the experimenter exchanged tapes in the tape recorder and remained s i l e n t u n t i l the 30 seconds lapsed. When 30 seconds lapsed, PT-1 then began (see below). 41 Post-Test Procedures Pos t - tes t 1 (PT-1) : Thi r ty seconds af ter the c h i l d l i s tened to the narrat ion the experimenter asked the c h i l d , while turning the tape recorder on to ' r e c o r d ' , "Would you please t e l l me the story you just heard? If you can not t e l l i t to me exact ly as you heard i t , then t e l l me the story as best as you c a n . " (The ra t iona le for the character of d i rec t ions i s given in Appendix B) . When the c h i l d was f in ished responding (or remained s i l e n t for 20 seconds), the experimenter s a i d , " Te l l me everything or anything you remember about the s tory . " When the c h i l d was f i n i s h e d , indicated ei ther by statement or by f a i l u r e to respond for 20 seconds, the experimenter s a i d , "I w i l l give you a moment to th ink . When I t e l l you, i f there i s anything else you would l i k e to t e l l me which you have not said yet , you may t e l l me then . " After a 30 second pause, the c h i l d was asked, "Is there anything else you remember about the story that you would l i k e to t e l l me?" When the c h i l d was f in ished responding, as indicated by saying so or through a 20 second period of s i l e n c e , PT-1 was terminated. However, before returning to the classroom, the c h i l d was t o l d , "Please .do not t e l l anyone what you did today." The experimenter then continued in two d i f f e r e n t ways. For the Y group the experimenter s a i d , " L e t ' s make t h i s our secret and not t e l l anyone e l s e , unless you wish to t e l l your parents ." For the 0 and M groups the experimenter s a i d , "It i s important for th i s study that you do not t e l l anyone what we did today. Of course 42 i f your parents ask you, you may t e l l them, but i t i s important that you do not t e l l any of your classmates. When I am f in ished with a l l of the ch i ldren who w i l l p a r t i c i p a t e I w i l l t e l l you. Then i f you wish to ta lk about what you did today you may. Can you do th i s for me?" After the c h i l d responded he or she was returned to h is or her c l a s s . This concluded P T - 1 . Post-Test 2 (PT-2): one week a f te r each c h i l d heard the nar ra t ion , he/she was again tested on his/her r e c a l l of the story . This was done in two par ts . Part One: af ter the c h i l d entered the tes t ing room, a br ief (one minute) warm-up or v i s i t intended to help put each c h i l d at ease occured. After the v i s i t the experimenter s tated, "Last week you heard a story on t h i s tape recorder ." While saying t h i s the experimenter pointed to the tape recorder and turned i t on. The experimenter continued: "Would you please t e l l me the story you heard on th i s tape recorder? If you can not t e l l i t to me exact ly as you heard i t , then t e l l me the story as best as you can . " The exact d i rec t ions as given in PT-1 were then followed to where PT-1 terminated. At t h i s point the f i r s t part of PT-2 terminated. Part Two: The second part of PT-2 began immediately a f ter the f i r s t part of PT-2 concluded. The experimenter s a i d , "I am going to ask you some questions about the s to ry . " The experimenter then read a set of questions for the c h i l d to answer (see Appendix C) . After a l l r e c a l l tasks were complete the Quick Test (QT) was administered. When the QT was completed, the experimenter concluded PT-2 by thanking the c h i l d 43 for coming in again , and asked the c h i l d not to t e l l his/her (class) f r iends about the test as was done at the end of PT-1. Moreover, prearrangements were made in that each c h i l d ' s teacher was asked to remind each c h i l d , p r i v a t e l y , not to discuss what had been done. Scoring Measures The purpose of t h i s study was to examine the extent to which four d i f f e r e n t encoding treatment condit ions affected memory performances of ch i ldren who var ied in age. Memory performance here had as i t s object of r e c a l l , narrat ive content which was i d e n t i c a l for a l l age-span groups and treatment condi t ions . Duration of the period allowed for encoding th i s content was also i d e n t i c a l for a l l subjects . With these and the other controls described in p lace, age and condit ion e f fec ts on narrat ive r e c a l l were assumed to be assessable. As indicated e a r l i e r , there are a number of researchers who are concerned with prose mater ial and i t s r e c a l l . However, these people appear to have been more concerned with how content features inf luence r e c a l l or the impact of the type of content used on r e c a l l ; and they have not studied how var ia t ions in age interact with encoding condit ions to inf luence r e c a l l . V a r i a b i l i t y in story c h a r a c t e r i s t i c s , such as length, content, theme development, and so for th are of in terest in such studies . As such, these story c h a r a c t e r i s t i c s are var ied while age and encoding condit ions are held constant. Thus, not a l l studies of 44 c h i l d r e n ' s memory of narrat ive prose examine age e f f e c t s . At the same time l i t t l e interest and at tent ion have been devoted to the study of the impact of encoding condit ions on r e c a l l of narrat ive content. For reasons of the same sor t , the present study keeps the story c h a r a c t e r i s t i c s (e .g . narrat ive content) constant while varying the age and encoding cond i t ions , the impact of which var ia t ions i s of in terest here. Therefore, in order to examine the inf luence of the encoding condit ions on the amount of r e c a l l , one passage was used for a l l age-span groups and a l l encoding treatment condit ions to maintain the same pattern of s p e c i f i c re la t ions across the experimental condit ions and age-span groups with regard to content. There are , of course, many ways of measuring and/or scoring free r e c a l l protocols from prose passages (see B a r t l e t t , 1932; Brown, 1975; Cofer , 1941; Mandler & Johnson, 1977; Meyer, 1975; Rubin, 1978a, 1978b; Rumelhart, 1975; Stein & Glenn, 1975b). However, what i s measured i s of concern. As Cofer (1941) has suggested, there are two major ways of examining r e c a l l mastery of prose. One e n t a i l s a focus on verbatim (or word for word) content, and the other focus i s on essent ia l ideas (idea content or g i s t ) of a passage. When d e s c r i p t i v e , argumentative, or expos i t ional discourse (three of the four types of connected discourse) are r e c a l l e d , exact or verbatim r e c a l l might be of concern since var ia t ions in r e c a l l of d e t a i l s from these types of discourse could be c r i t i c a l to the meaning or substance of them. In such cases, the f i r s t major way of examining mastery 45 of prose (focus on verbatim content) , might be more des i rable as a measure for these types of d iscourse . However, with the fourth type of connected discourse, narrat ion (which was used in t h i s study), var ia t ions in reconstruct ions of content do not necessar i ly change meaning or substance. Furthermore, the age and encoding condit ions enta i led in t h i s study may inf luence the degree to which reconstructions of the narrat ive depart from verbatim r e c a l l or are transformations or reconstruct ions of g i s t . In any event, minor va r ia t ions are commonly reported where subjects are asked to r e c a l l prose mater ia l ( B a r t l e t t , 1932; Brown, 1975; Meyer, 1975; Northway, 1940; Stein & Glenn, 1975a). As B a r t l e t t (1932) has reported, r e c a l l of prose may be changed to f i t one's "schema" when r e c a l l i n g a s tory . And, even i f "exact" r e c a l l i s required or asked fo r , transformations are normal rather than anomalous (Nelson, 1981) in reconstructed s t o r i e s . Since minor va r ia t ions must be expected, scor ing procedures should r e f l e c t t h i s . Furthermore, where ch i ld ren frequently resign or say less than they know when faced with what they regard as an impossible task (such as making an "exact" reproduction) , avoidance of t h i s i s des i rable i f one wants to avoid fa lse negatives. Accordingly , one does not ask for exact reproduction, and hence t h i s j u s t i f i e s adoption of a scoring system which r e f l e c t s the fact that exact reproduction was not requested. Thus, t h i s study used the l a t t e r focus or second major way of examining r e c a l l of prose. To repeat, the 46 present focus was on essent ia l ideas (or g i s t ) of the story , not verbatim r e c a l l . Therefore, two ways of measuring free r e c a l l were used. Both of these were designed to measure the essent ia l ideas or substance of the story , and take into account subst i tut ions and/or transformations of the story content (see below) . Another way of measuring memory of discourse i s through in te r rogat ion . One asks questions about the subject matter. Such interrogat ion may produce higher r e c a l l performances than free r e c a l l because the respondent i s cued by the questions. However, the age and/or treatment condit ions again may af fect these performances. Accordingly, an explorat ion of r e c a l l through interrogat ion may reveal possib le d i f f e r e n t i a l e f fec ts which may or may no't be evident in the analys is of free r e c a l l output. This a lso may add weight to the conclusions that can be drawn from analys is of the data. The foregoing considerations led to the adoption of three measures of the c h i l d r e n ' s r e c a l l of the story content. The f i r s t two of these were designed to measure the c h i l d r e n ' s free r e c a l l , and the t h i r d was designed to measure (test) the c h i l d r e n ' s remembrance of the story content as e l i c i t e d through quest ioning. It i s possible to use only one measure to explore the predict ions and the hypotheses in t h i s study. However, the use of three measures (as discussed below) allows for a more thorough examination of observed d i f fe rences , and provides a broader basis for drawing conclusions about the re la t ionsh ips 47 between the var iab les of interest than i s provided by a s ing le measure. Rubin's uni ts of measurement. The f i r s t instrument (or scoring scheme) employed was designed to measure free r e c a l l of s p e c i f i c word uni ts of a story as defined and described by Rubin (1978b). This measure scored s p e c i f i c "word uni ts" (units) of the text r e c a l l e d . This procedure allowed minor var ia t ions of the fo l lowing so r t : tense, case, number, and semantics. Noun to pronoun changes, and synonym subst i tu t ions are also acceptable. Therefore, t h i s measurement scheme allowed for subst i tut ions of the words which comply with the rules for the defined uni ts in the text in order to measure amount of content remembered but r e c a l l e d in a transformed s t a t e . At the same time i t maintains the same scoring procedure for a l l chi ldren in a l l cond i t ions . The story used in t h i s study contained 76 possible "un i ts" (see Appendix D) that could be r e c a l l e d . Rumelhart's story categor ies . The second instrument (or scoring scheme) employed was designed to measure free r e c a l l of the substance or e s s e n t i a l ideas of the t e x t . This measure was based on the "category" structure of a story developed by Rumelhart (1975) and modif ied by Stein and Glenn (1975b). This design separates a story into content categor ies . These categories or "schemas" denote se t t ings , events, in te rna l responses (thoughts and f e e l i n g s ) , 48 a c t i v i t i e s (excluding thoughts), and consequences. The story was categorized into Rumelhart's (Stein & Glenn, 1975b) schemas, or meaning u n i t s . This method, then, measures amount of r e c a l l by story category rather than by word units for ana lys is of the substance of the content. It al lows for g i s t to be counted in the scoring as wel l as the verbatim content. Since the category substance or e s s e n t i a l idea i s the concern for t h i s scoring procedure, the subjects are not penalized for transformed accounts which maintain the g i s t with f i d e l i t y . There were 41 "categories" (see Appendix E) derived from the s tory . Both of the aforementioned measures examine the essent ia l ideas or substance of a tex t . The word unit measure examines s p e c i f i c word uni ts which are based on "content" words described by Rubin (1 978b')_. The category measure examines s p e c i f i c phrases (categories) which include these "content" words as wel l as " funct ion" words omitted by Rubin (1978b). The word unit measure i s a smaller unit of measure than the category measure; and the category measure includes the contents of word un i ts but t reats same in terms of the category re ferents . However, i t i s possible to r e c a l l a word unit and not the category of a tex t , as wel l as i t i s possible to r e c a l l the e s s e n t i a l idea or substance of a category without the defined word u n i t . Therefore, for a more complete analys is of the free r e c a l l the use of both these measures i s des i rable in order to reveal any possible d i f f e r e n t i a l e f fec ts which may or may not be evident with one measurement of the free r e c a l l output. 49 Memory under interrogat ion (or cued r e c a l l measurement). A t h i r d means of measuring story r e c a l l was made by asking questions about the story . The questions (referred to as cued r e c a l l ) were a sample of poss ib le questions about the story content. These questions were used to allow for further ana lys is and comparison of the treatment cond i t ions . They were constructed in accordance with Spencer's (1973, c . f . Meyer, 1975) prose ana lys is system based upon who, what, why, when, where, and how quest ions. The interrogat ion of ch i ldren for r e c a l l of story content took place af ter the second free r e c a l l p o s t - t e s t . They were not asked af ter the f i r s t post - tes t so that the exchange enta i led at PT-1 would not confound the free r e c a l l resu l t s , at PT-2 (see F l a v e l l , 1977; K a i l , 1979; Kobasigawa, 1977; Meyer, 1979) A t o t a l of 24 questions were asked. The las t f i ve of these questions were not used as part of the t o t a l cued r e c a l l score since they were "opinion" quest ions. As such, there were 19 questions to be answered for scoring purposes. These are l i s t e d in Appendix C. Scoring procedures The measurement procedures y ie lded three sets of scores of r e c a l l data . These scores were derived from the three dependent measures: word u n i t s , story categor ies , and answers to questions (cued r e c a l l ) . These measures were scored as fo l lows : 1) Word unit scores were t a l l i e d using a predetermined l i s t made from the words derived from the story (see Appendix D). 50 Subst i tut ions of any word unit were counted as correct i f acceptable under the given ru les . 2) Category scores were t a l l i e d from a predetermined l i s t of categories (see Appendix E ) . A l l transformations of each category were i n d i v i d u a l l y judged or rated by two trained judges (raters) to determine whether or not the transformed category should be counted as cor rec t . Where disagreements arose, the raters were brought together to come to a mutual agreement. This procedure resulted in 100% agreement between the r a t e r s . 3) The answers given to the questions (from interrogat ion about the story at the end of the PT-2 free r e c a l l session) were t a l l i e d for correctness from a l i s t of predetermined answers (see Appendix C) . Since e i ther one does or does not r e c a l l a given instance of information, the data (word unit scores, category scores, and cued r e c a l l scores) were treated as nominal data and r e c a l l of same was scored as one point while f a i l u r e or incorrect r e c a l l was indicated by zero. Each c h i l d accumulated two scores for each of the word unit and category measures. Thus one pair was acquired at PT-1 and one pair was gained at PT-2. In a d d i t i o n , each c h i l d acquired one score from the number of correct answers on the cued r e c a l l measure (given as the second part of PT-2) . This comprised the scoring of the three measures (word unit r e c a l l , category r e c a l l , and cued r e c a l l ) of memory for the narrat ive content in the story . 51 Analys is Since t h i s study was a two factor (one c l a s s i f i c a t i o n , i . e . age factor p lus one treatment i . e . encoding factor) experiment (Ferguson, 1981), a f a c t o r i a l design with 3 leve l s of age and 4 of encoding condit ions was devised to organize observations of the e f f e c t s of these factors on the dependent (memory) measures. The p r i n c i p a l factors were age (Y, M, and 0) and treatment condi t ion (EC, IC, SC, and SIRC). A mul t i var ia te analys is of variance (MANOVA) was used to test for age and treatment e f f e c t s , as wel l as for age by treatment in te ract ion e f f e c t s . The c r i t e r i o n for judgement of s t a t i s t i c a l s ign i f i cance was p_<.05. For ind i v idua l age-span d i f fe rences , contrasts between the condit ions (within age-spans) were made to test for between group d i f fe rences . Tukey's HSD procedure was a lso used in t h i s study for post-hoc analys is which usual ly included a l l pairwise comparisons that could be made. Moreover, when making a l l poss ib le comparisons between means, such comparisons could not be regarded as independent. Therefore, t h i s post-hoc test was taken to be appropriate (Hays, 1973; K i r k , 1967). 52 CHAPTER IV Results The present study was designed to discover i f r e c a l l of narrat ive information i s af fected by age and encoding condit ions ( in which ch i ld ren encountered narrat ive m a t e r i a l ) . S p e c i f i c a l l y , t h i s study attempted to test hypotheses about the e f fec ts of condit ions (as stated in Chapter 2) within age-span groups and of age within cond i t ions . In order to test the hypotheses, a mul t i var ia te analys is of variance tested the f i r s t set of hypotheses raised in Chapter 2. This analys is tested for age by treatment i n t e r a c t i o n , age e f f e c t s , and treatment e f f e c t s . These resu l ts are reported in part A of t h i s chapter. For the second set of hypotheses raised in Chapter 2, contrasts between the d i f f e r e n t condit ion groups for the Y and 0 age groups tested for between condit ion e f fec ts within each age-span; and a one way analys is of variance test for d i f ferences between the condit ion groups within the M age-span was a lso undertaken as an exploratory a n a l y s i s . When d i f ferences were indicated by t h i s l a t t e r a n a l y s i s , pairwise contrasts between the groups were made in order to f ind where the d i f ferences were. These resu l t s are reported in part B of t h i s chapter. 53 A t h i r d sect ion (part C) i s included in t h i s chapter which gives resu l t s of a post-hoc a n a l y s i s . The purpose of t h i s analys is was to explore empir ica l connections between the encoding treatment condit ions and memory for narrat ive content performance. Since th i s analys is was of a decidely exploratory nature, the opportunity to compare resu l ts between the three age-spans provides some insight as to possible performance di f ferences not seen in the analys is in part A or B (and questions which they ra i sed ) . S t a t i s t i c a l contro l of v a r i a b i l i t y on the r e c a l l measures a t t r ibu tab le to other than the independent v a r i a b l e s , age and encoding cond i t ions , was sought through comparisons of a l l c e l l s with respect to gender, IQ ( a b i l i t y ) , and SES. These comparisons are reported at the end of th i s chapter (below, pages 78-79) . Analys is of resu l ts pert inent to the study now fo l lows. Part A Age and Treatment e f fec ts The prel iminary mul t i var ia te analys is of variance indicated that there was no age by encoding condit ion i n t e r a c t i o n . Therefore, hypothesis 1, which predicted age and encoding in teract ion e f f e c t s , was re jected . However, t h i s analys is d id reveal a s i g n i f i c a n t main ef fect of age, F(10,308)=13.17, 2<.001. As w e l l , there was a s i g n i f i c a n t main ef fect of treatment, F(15,426)=4.16, p^* 0 0 0 1 * Since the hypotheses were 54 concerned with both PT-1 and PT-2, the resu l ts from an analys is at PT-1 and PT-2 fo l low. Further , since the category measure was found to be highly cor re lated to the unit measure (a co r re la t ion c o e f f i c i e n t of .93 between the category and unit measure at PT -1 ; and a co r re la t ion c o e f f i c i e n t of .94 between the category and unit measure at PT-2) the unit measure performances were dropped from further considerat ion since they contr ibute l i t t l e that i s not merely redundant for present purposes. Hereafter , the measures under discussion should be understood to be the category measures unless otherwise stated. Moreover, since i t i s the univar iate resu l ts that are of interest (seen espec ia l l y at PT-2) Bonferroni 's procedure (K i rk , 1968, pp. 79-80) of d i v id ing the s ign i f i cance leve l for hypothesis tes t ing was used. Therefore, the c r i t i c a l region of acceptance for the hypotheses tes t ing was reset to p_<.01. An univar iate analys is of variance at PT-1 revealed a s i g n i f i c a n t main ef fect of age, F (2 , 1 58) = 61 . 41 , p/^OOO1? a n ( 3 a s i g n i f i c a n t main ef fect of treatment, F (2 , 1 58 ) =6. 08 , p_<.0007. At PT-2 there was again a s i g n i f i c a n t main ef fect of age (F (2 , 1 58 ) =41 . 62 , p_<.000l; for the cued r e c a l l measure, F(2,158)=36.79, 2 < « 0 0 0 1 ) - A s w e l l , at PT-2, there was a s i g n i f i c a n t main ef fect of treatment (F (3 ,1 58) =9. 03 , p_<.000l; and for the cued r e c a l l measure, F (3 , 1 58 ) = 1 9. 26, p_<« n00l). Nonorthogonal contrasts between the age-spans across the treatment condit ions indicated that at PT-1 the Y age-span was s i g n i f i c a n t l y i n f e r i o r to the M age-span (F(1,158)=13.34, 55 2<.0004), and to the 0 age-span (F (1 , 1 58) = 1 09. 48 , p_<.000l). The M and 0 age-spans were not s i g n i f i c a n t l y d i f fe ren t at PT -1 . Therefore, hypothesis 2, which predicted superior r e c a l l by the oldest age-span (as compared to the Y and M age-spans), was re jected ; while hypothesis 3, which predicted superior r e c a l l by the M age-span (as compared to the youngest age-span), was supported at PT -1 . At PT-2 the mean performances for the Y and M age-spans were not s i g n i f i c a n t l y d i f fe rent from each other . However, the Y age-span was s i g n i f i c a n t l y i n f e r i o r to the O' age-span (F(1,158)=71.97, £<.0001; and on the cued r e c a l l measure, F ( 1 , 1 58 ) = 51 . 37 , p_<.000l). As w e l l , the M age span was also found to be s i g n i f i c a n t l y i n f e r i o r to the 0 age-span (F (1 , 1 58 ) = 1 1 . 50 , p_<.000l; a n < ^ o n t n e cued r e c a l l measure, F(1,158)=22.20, £><.0001). Accordingly , at PT-2 hypothesis 2 was supported while hypothesis 3 was re jected . For the treatment condit ions (across the ages), at PT-1 the IC condit ion group was found to be superior to the SC group (F (1 , 1 58) = 1 5. 01 , p_<.0002). S i m i l a r l y the IC condit ion group was found to be superior to the SIRC group (F (1 , 1 58) =9. 46, p_<.0025). No other d i f ferences were found at PT -1 . Accordingly , hypothesis 4, which predicted r e c a l l d i f ferences between the EC and IC condit ions was rejected at PT -1 . Further , hypothesis 5, which predicted r e c a l l d i f ferences between the EC and both symbolic condit ions (SC and SIRC), was also rejected at PT -1 . 56 At PT-2, the EC condit ion group was found to be superior to the SC group (F(1,158) = 17 . 42, p ^ - 0 0 0 1 ) * 0 n t h e c u e d r e c a l l measure, F ( 1 , ,1 58 ) = 35 . 86 , p<.0001. The EC condit ion group was also found to be superior to the SIRC group (F(1,158)=15.11, p_<.0002; on the cued r e c a l l measure, F (1 , 1 58) = 29. 34 , p ^ ' 0 0 0 1 ) * Further, the IC condit ion group was found to be superior to the SC group (F(1 , 1 58) = 1 3.99, p_<.0003; on the cued r e c a l l measure, F(1,158)=32.96, p_<.0001). A l s o , the IC condit ion group was found to be superior to the SIRC group (F (1 , 1 58 ) =9 . 57 , p_<.0024. The same • resu l t was found on the cued r e c a l l measure, F (1 , 1 58) = 21 . 90 , p_<.000l). Therefore, hypothesis 4 was rejected while hypothesis 5 was supported at PT-2. Part B . Within Age-Span Comparisons Between the Conditions Again, for c l a r i t y and s i m p l i c i t y , only the PT-1 category measure and PT-2 category and cued r e c a l l measures performance resu l ts are reported (given the high co r re la t ion c o e f f i c i e n t s obtained between the category and unit measures at PT-1 and at PT-2 as reported above). Figure 2A (below, page 57) i s a graphic representation of the mean r e c a l l performances on the category measure for each age-span group in each condit ion at PT-1 and PT-2. A graphic representation of performances on the cued r e c a l l measure (PT-2) can be seen in Appendix F, Figure 2B. The analys is of resu l ts pert inent to each age-span group fo l lows : 57 Figure 2A Mean r e c a l l performance (PT-1 1 and PT-2 2) on the category measure by each treatment condition group (EC, IC, SC, and SIRC conditions) for each age-span group M 23 E 22 A 21 N 20 19 P 18 E 17 R 1 6 F 15 0 1 4 R .1 3 M 12 A 1 1 N 10 C 09 E 08 07 S 06 C 05 0 04 R 03 E 02 S 01 00 -+ + +-E I S c c c Treatment Conditions for Y Age-Span -+ + + + +-S E I S S I C C C I R R C C Treatment Conditions for M Age-Span -+-E C -+-I c -+-s c - + s I R c Treatment Conditions for O Age-Span 1- The upper most extremity (.) in each condition represents the mean performance on the category measure at PT-1. 2 _ The bottom most extremity (.) in each condition represents the mean performance on the category measure at PT-2. The v e r t i c a l bar (|) between the two extremities represents the decrement from PT-1 to PT-2 on the category measure for the conditions. 58 Y age-span comparisons (between cond i t ions ) . For the performance on a l l measures (PT-1 and PT-2) by the Y age-span group, the means and standard dev iat ions for each condit ion are presented in Table III (below, page 59). A graphic representation of the means on each measure for each condit ion group at PT-1 and PT-2 i s presented in Figure 3A and 3B (below, page 60). An i n i t i a l univar iate analys is of var iance contrast (Helmert contrast) indicated that there was no s i g n i f i c a n t d i f ference between the EC condit ion and the other condit ions when contrasted at PT-1 and PT-2. Moreover, t h i s was case on the cued r e c a l l measure too. Therefore, hypothesis 6, which s t ipu la ted that the EC group would-be superior on performance to a l l the other condit ions (IC, SC, and SIRC) was re jected. However, a post-hoc analys is revealed that c h i l d r e n in the EC condit ion were s i g n i f i c a n t l y superior to the ch i ld ren in the SC condit ion at PT-1 and on the category and cued r e c a l l measures at PT-2 (Tukey HSD, p_<.05). As w e l l , the EC group was s i g n i f i c a n t l y superior on performance when compared to the SIRC group on the cued r e c a l l measure at PT-2 (Tukey HSD, p_<.05). However, the IC group was s i g n i f i c a n t l y superior to the SC group at PT-1 and on the PT-2 cued r e c a l l measure. A l s o , the IC group was found, to be s i g n i f i c a n t l y superior to the SIRC group on the cued r e c a l l measure at PT-2 (Tukey HSD, 2<.05). I t appears that the IC group was accounting for the nons ign i f icant d i f ferences found when tes t ing th i s hypothesis. This i s evident since at 59 Table III Means and standard deviat ions of each measure at PT-1 and PT-2 by the Y age-span group in each condit ion PT-1 PT-2 Measure Condition Mean SD. Mean SD. Category: EC 6.53 3.81 5.40 3.66 IC 7.67 4.42 4.60 3.02 SC 2.80 2.31 2.33 1.84 SIRC 4.21 3.40 3.00 2.75 Mean Total 5.32 3.98 3.84 3.08 Un i t : EC 13.47 7.02 10.53 6.19 IC 15.00 7.72 9.06 • 4.64 SC 6.13 4.45 4.47 3.40 SIRC 9.50' 6.25 5.71 4.01 Mean Total 11.50 7.22 7.47 5.19 Cued R e c a l l : EC 8.60 2.59 IC 7.07 2.28 SC 4.33 2.32 SIRC 4.21 2.39 Mean Total 6.08 3.00 n: EC=15; IC=15; SC=15; SIRC=14; Total= 59. both PT-1 and PT-2 the mean r e c a l l scores for the IC group c lose l y resembled those of the EC group. A summary of these resu l t s i s presented in Table IV (below, page 61). 60 Figure 3A Condition group performance means for Y age-span (category and unit measures at PT-1) M E A N S C o R E S 17 16 15 14 13 12 1 1 10 09 08 07 06 05 04 03 02 01 00 Legend: category= unit = V -+ + + + EC IC SC SIRC Condition Groups (Y Age-Span) Figure 3B Condition group performance means for Y age-span (category, u n i t , and cued r e c a l l measures at PT-2) M E A N S C 0 R E S 16 15 14 13 12 1 1 10 09 08 07 06 05 04 03 02 01 00 Legend: category: unit = cued= -+ + + + EC IC SC SIRC Condition Groups (Y Age-Span) 61 Table IV Summary of the resu l ts of a l l pa i r wise comparisons on the PT-1 category and PT-2 category and cued r e c a l l measures (Y age-span) Group Measures PT-1 PT-2 Contrast Category Category Cued Reca l l EC vs IC EC vs SC EC vs SIRC IC vs SC IC vs SIRC SC vs SIRC * = s i g n i f i c a n t at p<.05 (Tukey HSD Test, Post -Hoc) . Absence of as te r i sks = no s i g n i f i c a n t d i f ference between groups M age-span group comparisons (between cond i t ions ) . For the performance by the M age-span group at PT-1 and PT-2 on a l l measures, the means and standard deviat ions for each condit ion are presented in Table V (below, page 62). Further , a graphic representation of the means on each measure for each condit ion group can be seen in Figures 4A and 4B (below, pages 63-64). 62 Table V Means and standard deviat ions of each measure at PT-1 and PT-2 by the M age-span group in each condit ion PT-1 PT-2 Measure Condition Mean SD. Mean SD. Category: EC 14.23 3.79 10.38 1.89 IC 14.00 5.16 8.35 3.71 SC 11.00 4.58 5.69 4.23 SIRC 9.86 4.67 5.93 3.20 Mean Total 12.26 4.81 7.57 3.80 Uni t : EC 23.92 4.91 17.08 3.84 IC 24.50 7.78 14.71 5.65 SC 20.46 6.75 10.77 7.38 SIRC 18.21 6.23 11.43 6.13 Mean Total 21.76 6.85 13.43. 6.28 Cued r e c a l l : EC '• 9.85 1.91 IC 9. 43 2.50 SC 5.69 3.12 SIRC 6.57 2.74 Mean Total 7.89 3.10 n: EC=13; IC=14; SC=13; SIRC=14; Total* 54. A one way univar iate ana lys is of variance on the category measure at PT-1 d id not indicate that there were any s i g n i f i c a n t d i f ferences between the condi t ions . However, at PT-2 s i g n i f i c a n t d i f ferences were revealed (F(3,50)=5.71, p<.002; and for the cued r e c a l l measure, F(3,50)=8.35, 2 < « ° 0 1 ) . Pairwise comparisons between the treatment groups on the measures using Tukey's HSD procedure indicated that at PT-2 the EC group was 63 Figure 4A Condition group performance means for M age-span (category and unit measures at PT-1) s i g n i f i c a n t l y superior to the SC and SIRC groups on both (category and cued r e c a l l ) measures (Tukey HSD, p_<.05). Further , the IC group was superior to the SC and SIRC groups on the cued r e c a l l measure. A summary table of the resu l t s can be seen in Table VI (below, page 65). 64 Figure 4B Condition group performance means for M age-span (category, un i t , and cued r e c a l l measures at PT-2) Condition Groups (M Age-Span) 65 Table VI Summary of the resu l ts of a l l pairwise comparisons on the PT-1 category measure and the PT-2 category and cued r e c a l l measures (M age-span) Condition Measures Group PT-1 PT-2 Contrast Category Category Cued Reca l l EC vs IC EC vs SC * * EC vs SIRC * * IC vs SC * IC vs SIRC - * SIRC vs SC absence of as te r i sk = no * = s i g n i f i c a n t at 2<'u5 s i g n i f i c a n t d i f ference between (Tukey HSD Test, Post-Hoc) groups. 0 age-span comparisons (between cond i t ions ) . The means and standard deviat ions of the measures performances for the 0 group condit ion subgroups at PT-1 and PT-2 are presented in Table VII (below, page 66). Furthermore, the means of each measure in each condit ion are graphica l l y presented in Figures 5A and 5B (below, pages 67-68) . Moreover, since i t i s obvious from close inspection of the mean r e c a l l performances that the SIRC group was not superior to at least one other condi t ion group at PT-1 or PT-2, as hypothesis 7 predicted , hypothesis- 7 was re jected. F i n a l l y , a post-hoc 66 analys is (using Tukey's HSD test) d id not reveal any s i g n i f i c a n t d i f ference between any of the groups on t h i s analys is at PT-1 or PT-1. Table VII Means and standard deviat ions of each measure at PT-1 and PT-2 by the 0 age- span group in each condit ion PT-1 PT-2 Measure Condition Mean SD. Mean SD. Category: EC 1 3.93 3.91 1 0.79 3.12 IC 1 5.29 5.31 1 1 .64 3.43 SC 1.2.80 5.33 8.67 5.45 SIRC 14.00 5.83 8.43 4.83 Total 1 3.98 5.09 9.86 4.44 Uni t : EC 24. 14 5.48 18.86 3. 16 IC 26.86 8.39 20.93 5.77 SC 22.73 8.59 1 5.00 8.30 SIRC 24.64 10.15 15.21 8.98 Total 24.56 8.24 17.46 7.24 Cued R e c a l l : EC 10.86 1 .99 IC 1 1 .57 2.59 SC 9.13 2.50 SIRC 9.36 3.73 Total 10.21 2.89 n: EC=14; 1 0 1 4 ; SC=15; SIRC=14; Total= 57. 67 F i g u r e 5A C o n d i t i o n group performance means f o r 0 age-span ( c a t e g o r y and u n i t measures a t PT-1) C o n d i t i o n Groups (0 Age-Span) 68 F i g u r e 5B C o n d i t i o n group performance means f o r 0 age-span ( c a t e g o r y , u n i t , and cued r e c a l l measures a t PT-2) C o n d i t i o n Groups (0 Age-Span) 69 Part C Other Observed Di f ferences : Exploratory Post-Hoc Analys is Observed d i f ferences which a r i s e from comparisons made for reasons other than those required for tes t ing the hypotheses set for th above are worthy of examination. Figure 6 (below, page 72) i s a graphic representation of the mean r e c a l l performances at PT-1 on the category measure across the age-span groups within each cond i t ion . A s imi la r graphic representation of the mean r e c a l l performances on the PT-2 category measure can be seen in Figure 7 (below, page 73). L ikewise, a graphic representation of the mean cued r e c a l l performances within each condit ion (at PT -2) . can be seen in Figure 8 (below, page 74). As can be seen from f igures 6 -8 , the EC condit ion performance graph appears to f l a t t e n (or be at a plateau) for the older age-spans, while the graphs of the other condit ions appear to steadly r i s e as age increases . While these graphs indicate an in teract ion (a c ross ing or in tersect ion of slopes formed from means), there was no age by treatment in teract ion found in the i n i t i a l analys is (part A ) . However, just as one should exercise caution when deal ing with in teract ion e f fec ts (Glass & Stanley, 1970), noninteract ion should a lso be treated with thoughtful care since "such f a i l u r e to see our i n t u i t i v e notions re f lec ted per fec t l y in our mathematical models i s a hazard (or r e a l i t y ) of an attempt to present the rea l world 70 m a t h e m a t i c a l l y " ( G l a s s & S t a n l e y , 1970, p. 410). A n o n s i g n i f i c a n t i n t e r a c t i o n e f f e c t would i n d i c a t e t h a t the performances ( w i t h i n c o n d i t i o n a c r o s s age-spans) i n the c o n d i t i o n s were s i m i l a r , or t h a t when w i t h i n c o n d i t i o n age-span means a r e p l o t t e d and compared t o the p l o t s of the o t h e r c o n d i t i o n s , they s h o u l d be p a r a l l e l . An i n s p e c t i o n of f i g u r e s 6-8, however, does not i n d i c a t e t h i s . These f i g u r e s evoke c u r i o s i t y about the adequacy of the ma t h e m a t i c a l a n a l y s i s . F u r t h e r a n a l y s i s of between groups d i f f e r e n c e s and t r e n d s i n the data seems w a r r a n t e d f o r both t h e o r e t i c a l and p r a c t i c a l r e a s o n s , s i n c e the performances of the c h i l d r e n w i t h i n each age-span d i f f e r e d (note between c o n d i t i o n s comparisons i n P a r t B above). These c o n s i d e r a t i o n s f a v o r e d an e x p l o r a t o r y a n a l y s i s which was undertaken t o answer the f o l l o w i n g q u e s t i o n : What degree of p o l y n o m i a l b e s t a p p r o x i m a t e s the form of the c u r v e which c o n n e c t s the group means f o r each c o n d i t i o n ( a c r o s s the age-spans)? In o t h e r words, what l e v e l of p o l y n o m i a l f u n c t i o n b e s t d e s c r i b e s the form of the group means (d a t a p o i n t s ) of the v a r i o u s e n c o d i n g c o n d i t i o n s ? To answer t h i s q u e s t i o n , t e s t s f o r l i n e a r t r e n d s i n each c o n d i t i o n was un d e r t a k e n . T e s t s f o r d e v i a t i o n from a l i n e a r t r e n d were a l s o u n d e r t a k e n . T e s t i n g f o r l i n e a r t r e n d s a c r o s s the age groups ( w i t h i n c o n d i t i o n ) r e v e a l e d t h a t t h r e e c o n d i t i o n s had s i g n i f i c a n t l i n e a r t r e n d s a t PT-1 (t = 4.13, p_<.0002 f o r the IC c o n d i t i o n ; t = 5.95, P<.0001 f o r ' the SC c o n d i t i o n ; and t = 5.98, p_<« 0 0 0 1 f o r t h e S I R C 7 1 cond i t ion ) . Further , one cond i t ion , the EC cond i t ion , deviated s i g n i f i c a n t l y from a l inear trend (t = 3 . 1 2 , p < . 0 0 3 5 ) , and indicated a quadratic t rend. On the PT - 2 category measure the same resu l ts were found: the IC, SC and SIRC condit ions had s i g n i f i c a n t l i n e a r trends and the EC condit ion deviated from a l inear trend ( t = 5 . 5 9 , p _ < . 0 0 0 l for the IC cond i t ion ; t = 6 . 9 2 , J 2 < . 0 0 0 1 for the SC cond i t ion ; t = 7 . 0 2 , 2 < - 0 0 0 1 f o r t h e S I R C cond i t ion ; and the EC condit ion t = 2 . 2 7 , p < . 0 2 9 for deviat ion from a l inear t rend) . For the PT - 2 cued r e c a l l measure no deviat ions from l inear trends were found. However, a l l the condit ions d id have a s i g n i f i c a n t l i near trend (t = 9 . 8 2 , p _ < . 0 0 0 l for the EC cond i t ion ; t = 4 . 9 3 , p _ < . 0 0 0 l for the IC condi t ion ; t = 6 . 6 8 , p _ < . 0 0 0 l for the SC cond i t ion ; and t = 7 . 2 8 , p _ < . 0 0 0 l for the SIRC cond i t ion ) . As indicated in Chapter 2 , one might wonder i f condit ions of encoding mit igate age di f ferences in r e c a l l . Accordingly, the fo l lowing questions were posed: ( 1 ) Do a l l condit ion groups of the Y age-span d i f f e r s i g n i f i c a n t l y from a l l the M and 0 age-span condit ion groups at P T - 1 ? And ( 2 ) are a l l condit ion groups of the Y age-span s i g n i f i c a n t l y d i f f e r e n t from a l l the 0 age-span condit ion groups at P T - 2 ? To answer these two quest ions, a post-hoc ana lys is using the Tukey test (Hays, 1 9 7 3 ; K i r k , 1 9 6 7 ) was employed to test for a l l possible di f ferences between a l l ages across the condit ions at PT-1 and P T - 2 . The resu l t s of the analys is on the category measure (PT-1 and P T - 2 ) and on the cued r e c a l l measure (PT - 2 ) can be seen in tables 72 VI I I -X (below pp. 75-77) . Figure 6 Performance means for each age-span group in each condit ion on the category measure at PT-1) Age-Span Groups 73 Figure 7 Performance means for each age-span group in each condit ion on the category measure at PT-2 M E A N P E R F 0 R M A N C E S C 0 R E S 27 26 25 24 23 22 21 20 19 18 1 7 1 6 1 5 14 13 12 1 1 10 09 08 07 06 05 04 03 02 01 00 Legend: EC = IC = SC = SIRC = Y M 0 Age-Span Groups 74 Figure 8 Performance means for each age-span group in each condit ion on the cued r e c a l l measure at PT-2) Table VIII Summary table of resu l ts of the Tukey HSD post-hoc analys is on the category measure (at PT-1) with a l l age-spans across a l l condit ions age Y Y Y Y M M O O M O M O condit ion S S E I S S S E I S E I C I C C I C C C C I C C R R R mean C C C 2.80 Y SC 4.22 Y SIRC 6.53 Y EC 7.67 Y IC 9.86 M SIRC * * 1 1 .00 M SC * * 1 2.80 0 SC * * * 13.93 O EC * * * * 1 4.00 M IC * * * * 14.00 O SIRC * * * * 1 4.23 M EC * * * * 15.29 O IC * * * * * = Denotes pa i rs of groups s i g n i f i c a n t l y d i f fe ren t at p_<.05. 76 Table IX Summary t a b l e of r e s u l t s of the Tukey HSD post-hoc a n a l y s i s on the c a t e g o r y measure (at PT-2) w i t h a l l age-spans a c r o s s a l l c o n d i t i o n s age Y Y Y Y M M M 0 0 M O 0 span c o n d i t i o n S S I E S S I s s E E I C I C C C I C I c C C c R R R mean C C C 2.33 Y SC 3.00 Y SIRC 4.60 Y IC 5.40 Y EC 5.69 M SC 5.93 M SIRC 8.36 M IC * * 8.43 0 SIRC * * 8.67 0 SC * * 10.38 M EC * * * * * 10.79 0 EC * * * * * * 1 1 .64 0 IC * * * * * * * = Denotes p a i r s of groups s i g n i f i c a n t l y d i f f e r e n t a t p_<.05. 77 Table X Summary table of resu l ts of the Tukey HSD post-hoc ana lys is on the cued r e c a l l measure (at PT-2) with a l l age-spans across a l l condit ions age Y Y M M Y Y O O M M O O span condit ion S S S S I E S S I E E I I C C I C C C I C C C C R R R mean C C C 4.21 Y SIRC 4.33 Y SC 5.69 M SC 6.57 M SIRC 7.07 Y IC 8.60 Y EC * * 9.13 0 SC * * * 9.36 0 SIRC * * * 9.43 M IC * * * 9.85 M EC * * * 10.86 0 EC * * * * * 1 1 .57 0 IC * * * * * * = Denotes pa i rs of groups s i g n i f i c a n t l y d i f fe ren t at p_<.05. 78 A b i l i t y ( I Q ) , SES, and Gender V a r i a b l e s  A b i l i t y ( I Q ) . A oneway a n a l y s i s of v a r i a n c e on the s c o r e s o b t a i n e d from the a b i l i t y f a c t o r i n d i c a t e d t h a t t h e r e were no s i g n i f i c a n t d i f f e r e n c e s between the t h r e e age t i e r s on t h i s measure. Moreover, no s i g n i f i c a n t d i f f e r e n c e s were found between the c o n d i t i o n groups i n r e g a r d t o the a b i l i t y measure w i t h i n each age-span. Gender and SES. The p r o p o r t i o n s of males and females i n the t h r e e ( a g e - t i e r ) d i s t r i b u t i o n s d i d not d i f f e r when they were compared w i t h a c h i square t e s t s t a t i s t i c ( S i e g e l , 1956). Nor d i d the a g e - t i e r s sub-samples d i f f e r i n t h e i r s o c i o - e c o n o m i c c l a s s c o m p o s i t i o n when the d i s t r i b u t i o n s of p r o p o r t i o n s f o r each of f o u r SES c a t e g o r i e s were compared w i t h a Kolmogorov-Smirnov goodness of f i t t e s t ( S i e g e l , 1956). Moreover, no s i g n i f i c a n t d i f f e r e n c e s i n the d i s t r i b u t i o n s of t h e s e v a r i a b l e s were found between the c o n d i t i o n groups. T h i s i s as e x p e c t e d , g i v e n the random assignment p r o c e d u r e . S i n c e t h e r e were no s i g n i f i c a n t d i f f e r e n c e s i n the a b i l i t y , gender, and SES f a c t o r s between any subgroups of the sample, i t was assumed t h a t these v a r i a b l e s were e v e n l y d i s t r i b u t e d t hroughout th e sample, and as such, d i d not appear t o have any s y s t e m a t i c i n f l u e n c e on the dependent measures. No f u r t h e r 79 s t a t i s t i c a l a n a l y s i s f o r c o n t r o l of th e s e v a r i a b l e s was war r a n t e d s i n c e sample s u b - s e t s d i d not d i f f e r from one another w i t h r e g a r d t o gender or sex mix, s o c i a l c l a s s , and measured a b i l i t y . R e l i a b i l i t y of the Free R e c a l l Measurement S i n c e measures of f r e e r e c a l l were taken at 30 seconds and one week f o l l o w i n g the c h i l d r e n ' s encounter w i t h the n a r r a t i v e , an assessment of the r e l i a b i l i t y of measurement was i n c o r p o r a t e d i n the s t u d y . C a l c u l a t i o n of Pearson product-moment c o r r e l a t i o n s f o r t e s t ( P T ~ 1 ) - r e t e s t ( P T - 2 ) r e l i a b i l i t y r e v e a l e d r e s p e c t a b l e c o r r e l a t i o n s (on the c a t e g o r y measure r= .80; n= 170.; p_< .001; and on the u n i t measure r= .79; n= 170; p_< .001). A c c o r d i n g l y , one may c o n c l u d e t h a t measurement e r r o r was v e r y modest and t h a t the measures were h i g h l y r e l i a b l e . W h i l e the v a l i d i t y of the measures was d i s c u s s e d e a r l i e r i n Chapter 3, f u r t h e r support i s a f f o r d e d the v a l i d i t y of the measures by the h i g h r e l i a b i l i t y o b t a i n e d . 80 CHAPTER V Di scussion As was indicated in the f i r s t two chapters, the present study was undertaken to obtain empir ical evidence pert inent to conjectures about the development of memory for meaningful d iscourse. In p a r t i c u l a r , i t was informed by suggestions (a) that such memory depends upon knowing; (b) that such knowing depends upon representat ion; (c) that said knowing and representation are a f fec ted by the nature of the representat ional capac i t ies of the knower and the condit ions under which the knowledge i s constructed; (d) that the knower's representat ional capac i t ies are transformed, augmented and enhanced as the knower matures (through t ime) ; and f i n a l l y (e) that the character and extent of memory for or r e c a l l of meaningful discourse i s a f fec ted by the condit ions under which the discourse information i s encoded and the degree to which the representat ional systems in which the information is encoded have matured. The successive emergence in the development of representat ional systems which Bruner (1964; 1966; 1973) descr ibed, provided the t h e o r e t i c a l basis for the formulation of developmental hypotheses about r e c a l l performances. In the present study, the manipulation of representat ional maturity was made poss ib le through (1) the use of a c r o s s - s e c t i o n a l design 81 wherein the r e c a l l performances of ch i ldren varying in age from 5 to 11 years were compared; (2) the manipulation of representations which were ordered through the systematic va r ia t ion of encoding condit ions (whereby the four sub-sets from any one of the three age t i e r s encoded the narrat ive content in four d i f fe ren t encoding condi t ions ; and each of the four encoding condit ions was i d e n t i c a l to those encountered by corresponding sample sub-sets in the other two age t i e r s ) . The said manipulations and other design features const i tuted a means by which the developmental hypotheses discussed above could be tested by experimental procedures. In e f f e c t , the study attempted to explore how the in teract ion of encoding condit ions with age-affected capac i t ies a f fec ts r e c a l l of narrat ive content. The d i f f e r e n t i a l e f fec ts of encoding condit ions were assessed within age t i e r s through comparisons of r e c a l l performances of the sub-sets of the age cohort formed through random assignment. The d i f f e r e n t i a l e f fec ts of increas ing ly mature representat ional capac i t ies were assessed through exploratory analys is of the mean performances from the three age t i e r s (with one another) within the condi t ions. As was seen in chapter IV (Results) a large amount of data was gathered and reported. Therefore, th i s chapter provides a summary of the f indings and evaluation of the hypotheses. In add i t i on , a l t e r n a t i v e explanations of the resu l ts are considered. Furthermore, the resu l ts are considered in r e l a t i o n 82 to a number of t h e o r e t i c a l issues and questions including those raised in the f i r s t sections of t h i s document. Thereafter, d i rec t ions for further research which a r i se from or are underscored by the present f indings are descr ibed. F i n a l l y , the impl icat ions the resu l ts have for i n s t r u c t i o n a l questions are explored. Evaluation of Hypotheses 1 . Part A: Age and treatment e f f e c t s . As seen from the i n i t i a l a n a l y s i s , commonly reported age di f ferences in r e c a l l were also found in t h i s study. An in terest ing observation however, i s the v a r i a b i l i t y in performances of the M age-span. While the M age-span had per-foxmances l i k e the 0 age-span on the immediate r e c a l l task, they performed more l i k e the Y age-span ch i ldren on the one week delayed r e c a l l task. These resu l t s seem to r e f l e c t the intermediate degree of maturity of the i r cognit ive structures and functions (compared with the younger and older subsamples). The i n i t i a l analys is a lso indicated that the performances of the ch i ld ren in the iconic condit ion were superior to the ch i ldren in the symbolic (SC and SIRC) condit ions at PT-1 and PT-2. At PT-2 the ch i ldren in the EC condit ion were also superior to the ch i ldren in the symbolic cond i t ions . L a s t l y , although there was an absence of a s i g n i f i c a n t in teract ion e f f e c t , there were between age-span within cond i t ion , as wel l as within age-span between cond i t ion , d i f fe rences . It appears that 83 the r e s u l t s o b t a i n e d from t h i s study were not p o w e r f u l enough t o g i v e s i g n i f i c a n t i n t e r a c t i o n e f f e c t s . T h i s , of c o u r s e , r a i s e s q u e s t i o n s about the e x t e n t of q u a l i t a t i v e changes i n c o g n i t i v e f u n c t i o n i n g d u r i n g the time span i n c o r p o r a t e d i n t h i s s t u d y . F u r t h e r d i s c u s s i o n on t h i s p o i n t i s p r e s e n t e d below. 2. P a r t B: c o n d i t i o n v a r i a b i l i t y w i t h i n age-span groups. A. Youngest (Y) age-span group. While i t was h y p o t h e s i z e d t h a t the same observed r e l a t i o n s h i p s as those r e p o r t e d by T r a v i s and White (1979) would be found w i t h the youngest (Y) age-span group, the ev i d e n c e from t h i s study i n d i c a t e s t h a t not o n l y was the e n a c t i v e c o n d i t i o n a s s o c i a t e d w i t h o p t i m a l r e c a l l , but the i c o n i c c.ondition was a s s o c i a t e d w i t h o p t i m a l r e c a l l as w e l l . T h i s was found on both the immediate and one week d e l a y r e c a l l when these r e c a l l p erformances were compared t o the r e c a l l performances of the sy m b o l i c c o n d i t i o n s (SC and SIRC). B. M i d d l e (M) age-span group. For the m i d d l e age-span group t h e r e was no p r e v i o u s t h e o r e t i c a l b a s i s f o r p r e d i c t i n g t h a t the e n c o d i n g c o n d i t i o n s would d i f f e r e n t i a l l y a f f e c t r e c a l l p e r f o r m a n c e s . For t h i s r e a s o n , an e x p l o r a t o r y a n a l y s i s between the e n c o d i n g c o n d i t i o n s was made f o r t h i s age-span. No r e c a l l d i f f e r e n c e s were found on the immediate r e c a l l , but d i f f e r e n c e s were found on the one week d e l a y r e c a l l . F u r t h e r , the M age-span r e s u l t s were s i m i l a r t o 84 the 0 age-span resu l ts (see below) at PT-1, and s imi la r to the Y age-span resu l t s at PT-2 - as was found on the i n i t i a l a n a l y s i s . That i s , at PT-2, the EC and IC condit ions were superior to the symbolic condit ions (SC and SIRC) on the one week delay r e c a l l but not on the immediate r e c a l l task. C. Oldest (0) age-span group. Superior e f f i c i e n c y of symbolic representation (as compared to enactive and iconic representation) was hypothesized for the oldest group (0 age-span group). However, no s i g n i f i c a n t d i f ferences between any of the encoding condit ions were obtained when both the immediate and one week delay r e c a l l performances were tes ted . 3. Part C: exploratory a n a l y s i s . While the trend analys is indicated that there i s the p o s s i b i l i t y that the enactive condi t ion was p o s s i b i l y deviat ing from a l inear trend (and that the other condit ions were not ) , i t must be remembered that the EC condit ion dev iat ion was not of s u f f i c i e n t magnitude to be s i g n i f i c a n t l y d i f fe ren t from the l inear trends of the other cond i t ions . However, i t does indicate that possib ly older ch i ldren (than those sampled in th i s study) should be tested to gain a better estimate of trends from a greater ( i . e . l e s s , truncated) age range. Further discussion on t h i s l a t t e r point i s resumed below. 85 4 . Summary evaluat ion. The mean r e c a l l performance of ch i ld ren of 9 years and less in . the enactive encoding condit ion was superior to the mean performances of ch i ldren in the SC and SIRC cond i t ions , as was the case in the Travis and White (1979) study. However, the iconic condit ion was no less powerful for the same age-spans (up to 9 years of age). In passing we may notice that the Travis and White (1979) sample was younger than the Y group in the present study. After age 9 the enactive and iconic condit ions were not associated with advantage or disadvantage on r e c a l l of a nar ra t i ve . For ch i ldren between 9 and 11 years of age (those in the oldest age - t i e r in t h i s invest igat ion) there were no d i f ferences in r e c a l l between the cond i t ions . The evidence gathered from th i s study (within each age-span between condit ion e f fec ts ) appears to indicate an emergence of symbolic processing e f f i c i e n c y . This e f f i c i e n c y was evident af ter 7 years of age when immediate r e c a l l was asked f o r . However, r e l a t i v e e f f i c i e n c y of retent ion over one week decl ined for youngsters under 9 years of age in the symbolic cond i t ions . This was evident by the superior r e c a l l performances of the Y and M age-span % groups in the EC and IC cond i t ions ; and the i r poorer performances in the symbolic cond i t ions . The Y and M age-span groups were not as p r o f i c i e n t in the i r r e c a l l of the narrat ive mater ia l as the 0 age-span group. They (Y and M age-span groups) may not have f u l l y developed the i r capac i t ies to ass imi la te information in symbolic systems, or have not 86 matured or developed a l te rnat i ve systems which are more economical and powerful than that which i s establ ished through motoric or iconic organization of the i r experiences. As such, i t appears that in the M and Y age-spans, knowledge constructed in enactive or iconic schemes was more enduring and reconstructable than the symbolic construct ions (alone or with rehearsa l ) . Perhaps the younger c h i l d r e n ' s (Y and M age-span groups) cognit ive functioning i s more dependent upon enactive and iconic schemas (see Anderson, Spi ro , & Anderson, 1978; Bower, Black, & Turner, 1979; Richardson, 1969) than i s the case for the older ch i ldren (0 age-span) who were general ly more adroi t in ass imi la t ing information to a l l three systems of representation (enactive, i c o n i c , and symbolic) . As age increased (from 5 to 11 years of age) the importance of the encoding condit ions decreased since the l a t t e r had no d i f f e r e n t i a l e f fec ts on r e c a l l in the eldest group. While there were no s i g n i f i c a n t between-condition d i f ferences in the 0 group, s i g n i f i c a n t between-condition d i f ferences were observed in the younger age ranges. As w i l l be discussed below, t h i s is perhaps the most in terest ing f ind ing in th i s study with respect to i t s impl icat ions for i n s t r u c t i o n . As can be seen from the r e s u l t s , the two questions raised on page twelve can now be given tentat ive answers in r e l a t i o n to the study: 1) The r e c a l l advantage that age advance seems to confer remained constant within and across the various encoding condit ions at each p o s t - t e s t . However, th i s consistency i s only 87 seen at each p o s t - t e s t . Within the M age-span, there was a s h i f t between the post - tes ts within and across the condi t ions . Performances resembled those of the 0 age-span at PT-1 and l i k e those of the Y age-span at PT-2. 2) The comparative r e c a l l advantage which the enactive encoding condit ion conferred on young ch i ldren in the Travis and White (1979) report d id not appear in the present study. However, both the enactive and iconic condit ions of encoding narrat ive mater ial were associated with r e c a l l super ior i ty up to 9 years of age in the present study. In f a c t , as the c h i l d r e n ' s ages increased in the present i n v e s t i g a t i o n , there were fewer r e c a l l d i f ferences between the var iab le condit ions (seen espec ia l l y at PT-2) . This - la t ter fact i s of t h e o r e t i c a l i n te res t . That i s , the fact of diminishing d i f f e r e n t i a l e f fec ts of encoding condit ions which accompany age increases (as observed) has a bearing on the question of how accurate i s the Brunerian conjecture about the order of emergence for representat ional systems. A br ief reconsiderat ion of Bruner's t h e o r e t i c a l speculat ion and the inferences from and extensions of the same speculat ion which are pert inent to the interpretat ion of the present r e s u l t s i s in order. Reca l l that according to Bruner (1964; 1966; 1973) the three representat ional systems (enact ive, i c o n i c , and symbolic) emerge in the l i s t e d order. Moreover, while rudiments of a l l three systems are usual ly evident by the end of the second year, the maturity of the f i r s t or enactive system i s more advanced 88 then, and for some time thera f te r , than i s the maturity of the iconic and symbolic systems. While l i t t l e more than th i s i s e x p l i c i t l y set for th by Bruner, one may conjecture, as was done in the present work, that the systems that emerge in order a f ter the enactive system, w i l l eventually surpass the enactive system in power, economy and e f f i c i e n c y with respect to the representation and organizat ion of information. One index of such power, economy and e f f i c i e n c y i s memory. However, since the duration of the period(s) which must elapse before the l a t e r emerging systems equal and then surpass the premier system was (and s t i l l i s ) unknown, ex t rapo lat ion , speculat ion , and inference were a l l one had to estimate when such developments would be evident. One can extrapolate from Bruner's (1966; 1973) discussions and speculate that one of the two la te r emerging systems (the iconic system) may account for the often remarked upon changes which are reportedly modal in the seventh or eighth year. Compared with the enactive mode i t has superior capac i t ies for representating information that i s action-independent such as can be i l l u s t r a t e d by states or q u a l i t i e s which inhere in things acted upon. That i s , notable gains in cogni t ive functions which have been reported regular ly as emerging during the seven to eight year period may r e f l e c t the emergence of a degree of maturity in the second or iconic system which makes possible feats of thought for which the act ion-based system is less s u i t e d . Accordingly , these expectations about the maturational 89 pattern for the iconic system was re f lec ted in the hypotheses of t h i s study. S i m i l a r l y , Bruner's wr i t ings , and those of others such as Piaget (1968; 1976), suggest that the f u l l flower of symbolic powers emerges s t i l l l a t e r . This suggested that hypotheses should r e f l e c t an expectation that maturity of the symbolic system would be evident in superior memory performances in the oldest age t i e r ( i . e . toward the end of latency per iod) . When one turns to the evidence produced by the present i n v e s t i g a t i o n , several matters pert inent to Bruner's theore t i ca l conjectures emerge. F i r s t , c l ea r - cu t and unequivocal confirmation of h is conjectures (about the order of emergence) and the impl icat ions drawn about the patterns of maturation for the three systems are not provided by th i s study. However, the evidence tends to support, more than refute , the general out l ine of Bruner's ideas. While the design does not permit an actual test of the conjecture about order of emergence, the superior power of the enactive system for the youngest subjects i s inferable from comparisons of r e c a l l performances between condit ions for the youngest group. This i s espec ia l l y compelling when the between groups comparisons for the middle and older groups reveal a diminishing advantage in r e c a l l for the enactive condit ion and emerging symbolic e f f i c i e n c y , as re f lec ted in convergence of condit ion e f f e c t s . However, notice must be taken of the r e c a l l 90 performances of the Y subjects in the iconic condit ion whose performances are not i n f e r i o r to the i r cohorts in the enactive cond i t ion . This suggests that the maturity of the iconic system i s more advanced by the ages represented in the youngest age t i e r than was expected. Moreover, the f u l l power of a mature symbolic system was not evident ( in superior r e c a l l ) as expected in those oldest (0) subjects in the symbolic cond i t ion . This in turn suggests that f u l l maturity of the symbolic system may emerge la te r than expected. This conclusion or in terpretat ion i s bolstered in a strongly suggestive, i f not compelling way by the slopes of the curves formed by jo in ing a l l data points of group means from the respective or corresponding tests across ages. Said curves, when used as a basis for extrapolat ion about when f u l l symbolic system power may be evident, suggests future invest igat ion of these matters with older subjects than were included in the present study. In sum, the d i f ferences in memory performances in the present study do suggest that the three representat ional systems reach maximum e f f i c i e n c y in an order that does not contradict what one would expect given Bruner's speculat ions. The older (0) ch i ldren were better able to r e c a l l the narrat ive in a symbolic encoding condit ion than were the younger (Y and M) c h i l d r e n . It therefore appears' that the symbolic representat ional system emerged las t (as Bruner conjectured) with those ch i ld ren in the present inves t iga t ion . If such representation systems emerge at d i f f e r e n t times in a p a r t i c u l a r 91 order (enactive then iconic then symbolic) , one might expect that in childhood the ontogeny of the system which appears f i r s t w i l l be more advanced than w i l l be the case with the second. L ikewise, the second system w i l l be more mature than the t h i r d , u n t i l a l l reach f u l l maturity (unless they mature at d i f fe rent r a t e s ) . Unfortunately t h i s matter i s not se t t led here. However, i f successive emergence of maturity in the three representat ional systems i s a warranted conclus ion, as suggested, then there are obvious s c i e n t i f i c as wel l as educational impl i ca t ions which are worthy of cons iderat ion . Discussion and Educational Impl icat ions of th i s Invest igat ion "Since the pioneer work of B a r t l e t t (1932) on remembering, there has been a tendency to t r i v i a l i z e human learning in experiments designed to discover fundamental general p r i n c i p l e s . " (Entwist le , 1976, p. 1) Success in showing mastery of subject matter in schools depends, in p a r t , on the capacity to r e c a l l the information content of the various forms of discourse (e .g . nar ra t ion , expos i t ion , d e s c r i p t i o n , and argumentation). The development of t h i s capacity i s of in terest to teachers and to scholars who are concerned with gaining understanding about r e c a l l or memory for the purpose of enhancing r e c a l l of important information. 92 Consideration of what i s enta i led in such r e c a l l has, apparently, led some scholars to the conclusion that , in sp i te of the very substant ia l amount of time and research e f f o r t on memory, l i t t l e of p r a c t i c a l consequence i s known about memory (Neisser, 1982; Scr ibner , 1984). Even so, Wickelgren (1981) has taken issue with th i s conclus ion. However, in doing so he focuses exc lus ive ly on what may be c a l l e d studies of micro-memory phenomena. Wickelgren (1981) claims that there are three temporal phases involved in memory s tud ies : learn ing , storage (consol idat ion and fo rget t ing ) , and r e t r i e v a l ( r e c a l l and recogni t ion) . Further, he d is t inguishes micro -s tudies from macro-studies . on the fo l lowing bas is : Mic ro -s tud ies are concerned with the learning of s ing le (or a small set of) assoc ia t ions , and encoding these s ingle (or small set of) chunks. Storage i s concerned with the consol idat ion and forget t ing of such small c e l l s of learned information; and r e t r i e v a l i s concerned with a s ing le elementary act of r e c a l l i n g or recognizing some unit of information (e .g . word, concept, p ropos i t ion ) . Macro-studies consist of such studies involv ing mult ip le r e c a l l , ordered r e c a l l , free r e c a l l , c r e a t i v i t y , problem s o l v i n g , and comprehension of large units of t e x t . This d i s t i n c t i o n between micro- and macro-studies of memory phenomena i s important when one considers Wickelgren's (1981) contention that s i g n i f i c a n t knowledge and understanding has been gathered with regard to memory. For in making h i s case, 93 Wickelgren, e x p l i c i t l y and exc lus i ve l y , l i m i t s h is argument to the considerat ion of micro -s tud ies . While many studies of memory may have been able to discover s i g n i f i c a n t knowledge of memory within the framework of micro- considerat ions, the concerns that informed the assessments of the c r i t i c s to whom Wickelgren (1981) reacted might wel l be rooted in the macro-world (which he excludes) . If micro -studies t e l l us l i t t l e of s ign i f i cance with regard to how developing memory re lates to macro-phenomena such as the information content of d iscourse, and t h i s i s what makes memory important for teachers and learners (Neisser, 1982), one has reason to address such matters. This i s not to say that micro -studies have been altogether or e n t i r e l y po in t less ; but they may t e l l us l i t t l e about memory in natural sett ings (Neisser, 1982). For however cumulative the resu l ts of micro -studies might be, they may never be able to t e l l us what we need to know about macro-memory. Consideration of t h i s las t point ar ises since the procedures employed in micro-studies exclude the study of macro-considerations, such as how the form and unity and in te rna l re la t ions of the content af fect free r e c a l l ; or how d i f f e r e n t degrees of cogni t ive maturity, encoding condi t ions , and the nature and coherence of discourse mater ial interact to inf luence r e c a l l . Therefore, the u t i l i t y or relevance of micro-memory research to educational questions and issues may be doubted. Answers to questions which are cent ra l to p r a c t i c a l 94 problems and issues such as " . . .how pupi ls use the i r own past experiences in meeting the present and the future" (Neisser, p. 12, 1982), are sought through the designs of macro-memory s tud ies . The Ebbinghaus t r a d i t i o n of studying "pure memory" that i so la tes i t s e l f from previous learning (D iS ib io , 1982) and possib ly future learn ing , divorces i t s e l f from, as Neisser (1982, p.12) would say, "natural cond i t ions" . As teacher concerns are focused on educational issues and questions that e n t a i l memory for meaningful information and understanding, and memory for what makes sense (as opposed to the interest in r e c a l l of nonsense of concern in the Ebbinghaus t r a d i t i o n ) , macro- considerations such as those l i s t e d above have to be incorporated, in memory studies i f the study of memory i s to have s ign i f i cance or funct ional value (Scribner, 1984). This i s espec ia l l y so where external and eco log ica l v a l i d i t i e s are concerned (D iS ib io , 1982; Hultsch & Hickey, 1978). Accordingly , one has reason to suggest that macro-memory studies approach the concerns of teachers who must deal with more p r a c t i c a l problems of memory better than do studies which invest igate " . . . m e n t a l functions in i s o l a t i o n from one a n o t h e r . . . " (Scr ibner, 1984, p. 2) ; for macro-memory studies are concerned with p r a c t i c a l r e a l i t y and u t i l i t y more than i s presently the case in micro-memory s tudies . As such, macro-studies may have more p r a c t i c a l (or use) value because of respect for mundane real ism and eco log ica l as wel l as content v a l i d i t i e s than that which derives from exclusive concern for 95 experimental real ism and t h e o r e t i c a l pur i ty (Travis , 1984; in p ress ) . It i s not d i f f i c u l t to understand why there i s a considerable body of l i t e r a t u r e which emphasizes micro-memory s tud ies . One only needs to look at the present study to comprehend why the Ebbinghaus experimental t r a d i t i o n has continued to be so popular; and why one t r i e s to contro l as many var iab les as poss ib le . Since d i f ferences on r e c a l l performance measures have been reported with var iab les l i k e age, gender, SES, a b i l i t y (and the l i k e ) , c e r t a i n l y some methodological cont ro l of such var iables i s d e s i r a b l e . Further , c o n t r o l l i n g for these var iab les can contr ibute to c l a r i f i c a t i o n of observed re la t ionsh ips and .advance t h e o r e t i c a l in terpretat ion of. exper imental - f ind ings . However, such var iab les are frequently d i f f i c u l t to contro l in classroom or other ext ra - laboratory research s e t t i n g s . Moreover, since "attempts to measure representat ional capacity of short term memory have been met with numerous d i f f i c u l i t i e s " (Rohwer & Dempster, 1977, p. 411), d i f f i c u l t i e s may be encountered when t r y ing to apply the resu l ts gained from micro-memory s tud ies , or t r y ing to general ize from micro- to macro-memory s i t u t a t i o n s (see Klapp, Marshburn, & Lester , 1983). This i s not to say that we have not gleaned any information from micro-memory studies that have educational u t i l i t y . Even Rohwer and Dempster (1977) confirm t h i s when they suggest that "Teachers should be wel l a d v i s e d . . . to consider the p o s s i b i l i t y 96 of memory f a i l u r e whenever they present a great deal of new m a t e r i a l . . . " (p. 414) (One wonders though, i f even our grandparents could not have t o l d us t h i s ! ) . However, th i s i s reported in r e l a t i o n to d i g i t span capacity l i m i t s , and not to representat ional capac i ty . It i s yet unknown how much representat ional capacity i s ava i lab le (see Rohwer & Dempster, 1977) when one deals with sentences or s to r ies with chi ldren of d i f fe rent ages. In any case, the evidence of the present macro-memory study i s consistent with the r e l i a b l e f indings in micro-memory research which show that increases in age or maturity seem to be accompanied by increases in memory prof ic iency (within the range of age studied here in ) . The impact . of age-affected factors on memory has. be§n of interest for a long time. That increase in age a f fec ts memory favorably in ear ly l i f e and adversely in la te l i f e i s wel l documented (Honsick, 1983; Pozdek & M i c h i l i , 1982). Furthermore, these observations of age-affected factors on memory are being supplemented and ref ined by work which focuses on p a r t i c u l a r c lasses of var iab les which are known to change with age increase (Honsick, 1983; McGraugh, 1983; Pozdek & M i c h i l i , 1982). For example, ch i ldren may develop or increasingly become more p l a n f u l , aware, and s t ra teg ic in the i r approach to problems, as we l l as the i r preparation for r e t r i e v a l (Cavanaugh & Borkowski, 1980; F l a v e l l , 1977; F l a v e l l & Wellman, 1977; Kreutsner, Leonard, & F l a v e l l , 1975; Yussen & Levy, 1975). Some developmental d i f ferences on r e c a l l performances have 97 been shown by the present study. As can be seen from the resu l ts of the exploratory analys is (Tables VTI I -X) , between-ages performance d i f ferences decreased when the r e c a l l of the older (M versus Y; 0 versus Y) ch i ldren in the symbolic condit ions was compared to the younger (Y) ch i ld ren who were in the EC and IC condi t ions . When between-age-tier comparisons were made, age di f ferences were found. However, when between-age-tier comparisons were made across a l l cond i t ions , age di f ferences diminished to the point of ins ign i f i cance in a few cases (see Tables V I I I -X ) . The present resu l ts are c e r t a i n l y not conclus ive . However, the fact of observed diminishing di f ferences as age increases i s provocative. Results are suggestive of the p o s s i b i l i t y that i f a ser ies of studies or invest igators cons is tent ly f ind a pattern of diminution of between-condition d i f ferences in r e c a l l performances as age increases, a noteworthy developmental phenomenon w i l l be estab l i shed . C l e a r l y , questions s t i l l must be raised about how encoding condit ions interact with age-affected cognit ive c h a r a c t e r i s t i c s to a f fec t memory. The extent of these e f fec ts can cancel often-seen age advantage in memory when r e c a l l performances of ch i ld ren as young as 5 to 7 years are compared with l i k e performances of ch i ldren 9 to 11 years. The exploratory evidence indicates that ch i ld ren aged 5 to 7 years, when they encode narrat ive content in condit ions which enable them to encode same in enactive and iconic systems, do not d i f f e r in 98 the i r r e c a l l performances from that of ch i ld ren aged 9 to 11 years (the 0 chi ldren) who encode the same mater ial in condit ions which require that the narrat ive content be encoded in symbolic systems. The Y ch i ld ren in these same two condit ions do not d i f f e r from the M ch i ld ren (aged 7-9 years) in the SC and SIRC condit ions e i t h e r . Accordingly, one may question genera l i zat ions about memory development which do not take account of the influence of encoding condit ions and suggest that further invest igat ion of these matters i s warranted. In p a r t i c u l a r , conclusions from past research, which as a rule only en ta i l ed symbolic encoding, may warrant reconsideration in the l i g h t of present f ind ings . While i t i s d i f f i c u l t to incorporate procedures which embody enactive and iconic encoding condit ions in studies of memory for d i g i t s , eve 's , or s ingle words (other than concrete words) in a very short time span, the present pattern of resu l ts suggest that general izat ions which have been drawn from such studies warrant rev is ion in the l i g h t of the encoding condit ion e f f e c t s seen here in . For example, Brown (1975), in o f f e r i n g a c o r o l l a r y to F l a v e l l ' s dictum on mnemonics and e f f i c i e n t task performance, stated that when no mnemonic strategy i s required for e f f i c i e n t performance of a task, "the task w i l l be r e l a t i v e l y insens i t i ve to developmental t rends ." (p. 110). The evidence of the present study i s not e n t i r e l y in accord with t h i s dictum. While i t i s accepted that c h i l d r e n ' s performances on r e c a l l measures general ly increases with age ( Jab lonsk i , 1974; Stein & 99 Glenn, 1976; Cavanaugh & Borkowski, 1980), some researchers have suggested that i t i s the task of verbal recounting that presents d i f f i c u l t i e s for young ch i ldren (Brown, 1975). Moreover, young ch i ld ren may be unaware of the r e l a t i o n s h i p between organization and r e c a l l , or , are less apt to organize the i r r e t r i e v a l in free r e c a l l as adults (Lange & G r i f f i t h , 1977). S i m i l a r l y , Bruner (1973) has claimed that , "The key to r e t r i e v a l i s o rgan i za t ion . " (p. 411). Well organized information appears to be learned more readi ly and remembered longer than information that i s not wel l organized (Bower, C la rk , Lesgold, & Wineberg, 1964). Accordingly , there i s reason to conjecture that in the present study the EC and IC encoding condit ions f a c i l i t a t e d organizat ion which was re f lec ted in higher mean performances for the Y and M groups in these condit ions compared with the i r cohorts in the SC and SIRC groups. Further , "mater ial that i s organized in terms of a person's own interest and cognit ive structure i s mater ia l that has the best chance of being access ib le in memory" (Bruner, 1973, p. 412). It i s therefore suggested that the enactive and iconic encoding condit ions of t h i s study may have both (1) engaged the c h i l d r e n ' s (Y and M age-span) i n t e r e s t s ; and (2) maximized e f f i c i e n t ass imi la t ion of the narrat ive content by the i r current cogni t ive structures (as seen in the r e s u l t s ) . This would support Bruner's (1973) contention that "ch i ld ren do best in recovering mater ia l t i e d together by the forms of mediation they most often use . " (p. 411). Therefore, 100 Bruner's suggestion of successive emergence of representation systems from enactive to iconic to symbolic forms of representation i s not contradicted by the present invest igat ion since the younger ch i ldren (Y and M age-spans) were more p r o f i c i e n t when r e c a l l i n g information af ter encoding in the enactive and iconic condi t ions ; and since the the older ch i ldren (0 age-span) were more p r o f i c i e n t than were the younger chi ldren when r e c a l l was tested a f te r they encoded under symbolic cond i t ions . Bruner (1964), Piaget (1976), and Vygotsky (1978) have a l l suggested that cognit ive funct ioning in young ch i ld ren i s more dependent upon enactive schemas than i s the case with older c h i l d r e n . They seem to suggest that older ch i ldren are general ly more adroi t in a s s i m i l a t i n g information to . f igurat i ve aspects of both iconic and symbolic representat ional schemas than are the younger c h i l d r e n . This suggests that there i s development of cognit ive c h a r a c t e r i s t i c s from enactive to iconic to symbolic representation encoding a b i l i t i e s . Moreover, such d i f ferences seem to be associated with memory performance d i f ferences (as suggested by the present study).. Accordingly , one might consider the meaning of these d i f ferences in re la t ion to i n s t r u c t i o n . Ins t ruc t iona l Considerations Past research has shown that young ch i ldren r e c a l l more information when encoding in enactive systems of representation 101 than when encoding in symbolic representation systems (Paris & Lindauer, 1975; Par is & Scot t , 1975; Travis & White, 1979). As w e l l , there are many reports which indicate superior r e c a l l performances when ch i ldren are required to do something in connection with presented mater ial (Danner & Taylor , 1973; Lev in , Ghatola, DeRose, Wi lder , & Norton, 1975; Lev in , Lesgold, Shimron, & Guttman, 1975; Lev in , McCabe, & Bender, 1975; Par is & Lindauer, 1976; Par is & Upton, 1976; Richardson, 1969; Rubin & Po l lack , 1969; S i l ve rn & Yawkey, 1977; Travis & White, 1979). These considerations underscore the d e s i r a b i l i t y of f inding out i f memory performance i s enhanced throughout childhood when information i s organized and represented by enact ion. At the present t ime, enactive and iconic representations of discourse information seem to be superior to symbolic representations for r e c a l l by ch i ld ren up to age nine. Var iat ions in encoding condit ions do not seem to have d i f f e r e n t i a l e f fec ts on r e c a l l of narrat ive content af ter 9 years of age. Accordingly , where teachers require the r e c a l l of (narrative) information by chi ldren (5-11 years of age), superior r e c a l l performances may be expected i f the teacher arranges for ch i ldren ( 5 - 9 years of age) to organize the mater ia l enact ively or i c o n i c a l l y . However, t h i s i s not to imply that a l l mater ia ls should be addressed in these ways. Symbolic organizat ion should also be encouraged since a wel l developed a b i l i t y to represent information in symbolic forms i s 1 02 h ighly advantageous, and i t s development probably depends upon e f f o r t f u l p ract ice (Horton & M i l l s , 1984). F u l l development of such a b i l i t i e s might be retarded or may not be aided and abetted as we l l as might be the case i f teachers r e s t r i c t the encoding condi t ions to those which are most advantageous for r e c a l l in the short term or immediate sense. After a l l , education implies gains in knowledge, a b i l i t i e s , and s e n s i b i l i t i e s . Such gains require accommodation or modif icat ion of ex i s t ing structures and such accommodation requires that learners augment present c a p a c i t i e s to the requirements or d i s c i p l i n e of impersonal r e a l i t y . The r e c a l l of ch i ldren between 9 - 1 1 years of age may s t i l l benefi t or be enhanced from enactive or iconic o rgan iza t ion . However, symbolic ordering at th i s age may produce superior performance resu l ts where the c h i l d i s requested to r e c a l l the given information. Again, l i k e the younger c h i l d r e n , t h i s should not imply s t r i c t use of only symbolic encoding organizat ion , but possib ly judic ious emphasis of i t . Moreover, since the present study c l e a r l y shows that increase in age is accompanied by decreases in the d i f f e r e n t i a l e f f e c t s of encoding condit ions on r e c a l l of discourse informat ion , another impl icat ion for ins t ruc t ion i s that perhaps teachers of pupi ls over 10 or 11 years of age (and who are otherwise l i k e those in the present study) need not be gravely concerned about such encoding condit ions as they a f fec t memory of the sort studied here. 103 L imi tat ions and Caveat The present study, l i k e a l l s tud ies , i s character ized by ce r ta in l i m i t a t i o n s which warrant d i scuss ion . Some of these are connected with (1) age sample cons iderat ions ; (2) the free r e c a l l procedure; and (3) the rehearsal condit ion r e s u l t s . Each of these are discussed in turn below. Age sample. As reported, the present study incorporated random assignment of ch i ldren within each age-span to form the four condit ion groups in each age-span t i e r . As one would expect, s t a t i s t i c a l analys is confirmed that age in each age-span was evenly d i s t r i b u t e d across condi t ions . However, the d i s t r i b u t i o n of ch i ldren across the three two-year age-spans was not e n t i r e l y uniform. Tests of homogenity of variance on the three spans indicated that the d i s t r i b u t i o n of the Y ch i ldren in the 5-7 age-span d i f f e r e d from the d i s t r i b u t i o n of the 0 ch i ld ren in the 9-11 age-span (Bart let t -Box F=6.007; p_<.0l4). A l l other comparisons on homogenity of variance between the age t i e r s were not s i g n i f i c a n t ( ind icat ing that the d i s t r i b u t i o n s of the age variances were not heterogeneous). Accordingly , one may conclude that comparisons between adjacent age t i e r s are comparisons of groups which are s i m i l a r l y d i s t r i b u t e d within the i r respective age-spans. Therefore, for purposes of the present study, there i s l i t t l e reason to consider d i f ferences in homogeneity of d i s t r i b u t i o n s between the t i e r s . 104 Free r e c a l l technique. The act ions and operat ional routines of many teachers imply that ch i ld ren are expected to meet requests for information at some point in the future. Chi ldren readi ly discern these i m p l i c i t messages about what i s expected and presumably are affected by them in varying degrees. As was indicated e a r l i e r in Chapter III ( in the procedure of introducing the story in the study), no c h i l d was t o l d beforehand that they would have to r e c a l l the story a f ter they l i s tened to i t . Since no ins t ruc t ions suggested that r e c a l l was imminent, t h i s absence of ins t ruc t ion may have influenced the older c h i l d r e n ' s (M and 0 age-span groups) s t rateg ies in the encoding and "storage" of same ( F l a v e l l , 1977, p. 209). As such, the r e c a l l performances obtained from the M and 0 age-span groups may have been higher, as i s implied by F l a v e l l (1977), had the ch i ldren been forewarned. However, the Y age-span performance scores may not have been so af fected since ch i ld ren of th i s age apparently do not employ mnemonic s t rateg ies in an t i c ipa t ion of r e c a l l to the extent that older ch i ldren do (Brown, 1975; F l a v e l l , 1977). In sum, these considerat ions suggest that the between age group contrasts in th i s study might underestimate age- re lated d i f ferences in r e c a l l performances. Rehearsal e f f e c t s . As stated e a r l i e r , the SIRC condit ion was included because ch i ldren in the enactive condit ion necessar i ly would have 105 rehearsed the substance of the story as they dramatized i t . Children in the IC and SC condit ions did not have to rehearse the m a t e r i a l . Therefore, th i s condit ion (SIRC) was included to get an estimate of the extent to which rehearsal inf luences the r e c a l l measures in the enactive treatment. However, at no time was the SIRC condit ion s i g n i f i c a n t l y d i f fe ren t from the SC condit ion within any age-span group. ' As w e l l , there were no s i g n i f i c a n t d i f ferences found between the EC and IC condit ions within each age-span group. Further, the between age-span comparisons in the SIRC condit ion y ie lded s i m i l a r resu l ts as found in the SC cond i t ion . Therefore, since there was no apparent contr ibut ion of rehearsal to the scores of the SIRC cond i t ion , no estimation of the putative ensconced rehearsal e f fects was made. These expected rehearsal e f f e c t s , since they are apparently n e g l i g i b l e , did not mater ia l i ze to any notable degree. However, why rehearsal did not provide any apparent increase in the measured r e c a l l i s of interest since rehearsal probably should have had some ef fect ( K a i l , 1979). Perhaps there was not enough time a l lo ted to rehearse the information. The story was narrated at 142 words per minute, and was one minute and t h i r t y seconds in durat ion. The rehearsal time given was t h i r t y seconds. This may mean that those ch i ld ren in the SIRC condit ion would need to rehearse the story at 284 words per minute for verbatim rehearsal of the ent i re s tory . Since l i s t e n i n g comprehension a b i l i t y apparently ranges from 75 to 175 1 06 words per minute (Brosk i , 1974) i t i s possible that the ch i ld ren who were to ld to rehearse (SIRC subjects) did not have enough time to rehearse the complete s tory . This , of course, could help explain the non -s ign i f i cant d i f ferences found between the SC and SIRC condit ions with in each age-span group. However, u n t i l further research can c l a r i f y what i s rehearsed when one rehearses a story (verbatim content, g i s t , or an episode of the story -see Horton & M i l l s , 1984), one can only speculate about why the SIRC condit ion d id not have greater performance scores than the SC cond i t ion , as might be expected (see K a i l , 1979). D i rect ions for Further Research In addit ion to the foregoing caut ions, some questions a r i s e from the present study which need further i n v e s t i g a t i o n . One may ask: 1) At what age(s) do the iconic encoding a b i l i t i e s begin to have e f f i c i e n c y in construct ing knowledge s imi la r to that of the enactive encoding a b i l i t i e s ? Studies which include ch i ld ren younger than those in the present study may answer t h i s quest i on . 2) When the r e c a l l performances of ch i ld ren who are older than those in the present study are studied, w i l l the a b i l i t y of those who encode narrat ive mater ia l in symbolic systems (and then r e c a l l same) be associated with r e c a l l advantage (compared to r e c a l l performances of the i r cohorts who encode in the a l t e r n a t i v e systems)? Studies which include ch i ldren older than those in the present study should answer t h i s quest ion. 107 3) Since the present study was designed to incorporate one form of discourse (narration) only , the same questions that were addressed by, as wel l as those which a r i se from, the present study should a lso be explored in r e l a t i o n to the other forms of discourse - expos i t ion , d e s c r i p t i o n , and argumentation. 4) Just as i t may be important to study how much is remembered, so may i t be just as important to study what i t i s that i s remembered and how much and what kind of content i s forgotten under var iab le encoding condit ions at d i f f e r e n t ages. The use of the strategy seen in the present invest igat ion could be incorporated for the study of condit ions of forget t ing (with the decrement l i n e magnitude, as seen in Figure 1, as the index of forget t ing for each cond i t ion ) . Conclusion The invest igat ion under d iscussion was designed to reduce our ignorance in one of the realms which encompasses macro-memory. Phenomena from that realm are s i g n i f i c a n t for learning and teaching because learners and teachers re ly so heavi ly on the various forms or types of discourse in the pursuit of educational ob jec t i ves . The resu l t s of the present invest igat ion can be interpreted for such impl icat ions as they might have for ins t ruc t ion and theory since the question of how encoding condit ions and age-affected c h a r a c t e r i s t i c s interact to inf luence r e c a l l of information in one type of discourse (narration) was enta i led in the present study. 108 It does appear that both encoding condit ions and age-affected c h a r a c t e r i s t i c s inf luence r e c a l l . This i s c l e a r l y evident in the present i n v e s t i g a t i o n . While a.more complete theroet ica l explanation for the present resu l ts requires further research as indicated by the-quest ions raised above, educators may take note that t h i s study has shown how age and encoding condit ions can a f fec t the r e c a l l of narrat ive information by chi ldren ranging in age from 5 to 11 years af ter encoding under several cond i t ions . The planned work for future explorat ions of r e c a l l of discourse information encoded under various condit ions should c l a r i f y the degree to which th i s genera l i zat ion has warrant. 109 BIBLIOGRAPHY Ammons, R .B . , & Ammons, C.H. (1962). The quick test (QT): p rov is iona l manual. Psychological Reports. Monograph Supplement 1-V1. Ammons, R .B . , & Ammons, C.H. (1979). Use and evaluat ion of the quick test (QT): p a r t i a l summary through October, 1979: I I . Reviews, theses, unpublished reports and papers. Psychological Reports, 45, 953-954. Anderson, R .C . , Sp i ro , R . J . , & Anderson, M.C. (1978). Schemata as sca f fo ld ing for the representation of information in connected discourse. American Educational Research Journal , 15, 433-440. Bandura, A . , & Je f fe r y , R.W. (1973). Role of symbolic coding and rehearsal processes in observat ional l ea rn ing . Journal  of Personal i ty and Soc ia l Psychology, 26, 122-130. B a r t l e t t , F.C. (1932). Remembering. Cambridge: Cambridge Univers i ty Press . Be l lezza , F .S . (1981). . Mnemonic -devices: c l a s s i f i c a t i o n , c h a r a c t e r i s t i c s , and c r i t e r i a . Review of Educational  Research, 51, 247-275. Bender, B.G. & Lev in , J . R . (1978). P ic tu res , imagery, and c h i l d r e n ' s prose l e a r n i n g . Journal of Educational Psychology, 70, 583-588. B i jou , S.W. (1976). C h i l d development: the basic stage of  early chi ldhood. New Jersey : P r e n t i c e - H a l l . Bower, G .H. , B lack, J . B . , & Turner, T . J . (1979). Sc r ip ts in memory for t e x t . Cognit ive Psychology, 11, 177-220. Bower, T.G.R. (1979). Human Development. San Francisco: W.H. Freeman and Company. Brooks, L.W. & Dansereau, D.F. (1983). Ef fects of s t ruc tura l schema t r a i n i n g and text organizat ion on expository prose processing. Journal of Educational Psychology, 75, 811-820. B rosk i , D.C. (1974). Auditory learners and comprehension of rate a l te red recordings (auditory learning monograph ser ies 3) . East Lansing, M i c h . : Michigan State U n i v e r s i t y . (ERIC Document Reproduction Service No. ED 102751) 1 10 Brown, A .L . (1975a). The development of memory. Knowing, knowing about knowing, and knowing how to know. In H.W. Reese (ed.) Advances in c h i l d development and  behavior (Vol . 10). New York: Academic Press, 103-152. Brown, A .L . (1975b). Recognit ion, reconstruct ion, and r e c a l l of narrat ive sequences by preoperational c h i l d r e n . Ch i ld  Development, 46, 156-166. Brown, A .L . (1978). Knowing when, where, and how to remember. A problem of metacognit ion. In R. Glaser (ed.) Advances in  Ins t ruc t iona l Psychology ( vo l . 1). New Jersey: Erlbaum, 1 978. Brown, A . S . , & Oxman, M. (1978). Learning through p a r t i c i p a t i o n : E f fec ts of involvement and a n t i c i p a t i o n of involvement. American Journal of Psychology, 91, 161-172. Bruner, J . S . (1964). The course of cognit ive growth. American  Psychologist , 19, 1-15. Bruner, J . S . (1966). Toward a theory of i n s t r u c t i o n . Cambridge, Mass. : Harvard Univers i ty Press. Bruner, J . S . (1973). Beyond the information given. New York: W.W. Norton & Company Inc. 316-412. Bruner, J . S . & Olson, D.R. (1977-78). Symbols and texts as tools of i n t e l l e c t . Interchange, 8(4) , 1-15. C e c i , S . J . , Caves, R .D. , & howe, M.J.A. (1981 ) . Ch i ld ren 's long-term memory for information that i s incongruous with the i r p r io r knowledge. B r i t i s h Journal of Psychology, 72, 443-450. C h i , M.T. (1976). Short term memory l i m i t a t i o n s in c h i l d r e n : capacity or processing d e f i c i t s ? Memory and Cognit ion, 4, 559-572. C ra ik , F . I .M . (1979). Human memory. Annual Review of  Psychology, 30, 63-102. Cofer, C.N. (1941). A comparison of l o g i c a l and verbatim learning of prose passages of d i f fe rent lengths. The  American Journal of Psychology, 54, 1-20. Danner, F.W., & Taylor , A.M. (1973). Integrated p ictures and r e l a t i o n a l imagery t r a i n i n g in c h i l d r e n ' s learn ing . Journal of Experimental Ch i ld Psychology, 16, 47-54. Davis, W.E. , & Dizzonne, M.F. (1970). Relat ionship between the QT and the WAIS. Psychological Reports, 20, 457-458. 111 Dempster, F.N. (1981). Memory span: sources of ind iv idua l and developmental d i f fe rences . Psychological B u l l e t i n , 89, 63-100. D iS ib io M. (1982). Memory for connected discourse: a c o n s t r u c t i v i s t view. Review of Educational Research, 52, 149-174. Drum, P.A. (1985). Retention of text information by grade, a b i l i t y , and study. Discourse Processes, 8, 21-52. Dunham, T .C . , & Lev in , J .R . (1979). Imagery ins t ruc t ions and young c h i l d r e n ' s prose lea rn ing : no evidence of "support". Comtemporary Educational Psychology, 4, 107-113. Ebbinghaus, H. (1964). Memory: a contr ibut ion to experimental  psychology. New York: Dover Pub l i ca t ions . E l k i n d , D. (1971). Ear ly childhood education. A Piaget ian perspect ive. National Elementary P r i n c i p a l , 51, 48-55. Entwis t le , N . J . , (1976). Symposium: Learning processes and s t ra teg ies -1 ( E d i t o r i a l Introduction - The verb ' to learn ' takes the accusat ive) . * B r i t i s h Journal of Educational  Psychology, 46, 1-3. Ferguson, G.A. (1981). S t a t i s t i c a l analys is in psychology and  education (5th Ed . ) . New York: McGraw H i l l , 222. F l a v e l l , J . H . (1963). The developmental psychology of Jean  Piaget• Pr inceton, New Jersey: Van Nostrand. F l a v e l l , J . H . (1970). Developmental studies of mediated memory. In H.W. Reese & L . P . L i p s i t t (Eds.) Advances in  c h i l d development (Vol . 5 ) . New York: Academic Press , 182-211 . F l a v e l l , J . H . (1971). F i r s t d iscussant 's comments: what i s memory development the development of? Human Development, J_4, 272-278. F l a v e l l , J . H . (1977). Cognit ive development. Englewood C l i f f s , New Jersey: P r e n t i c e - H a l l , Inc . , 56-87; 198-216. Fur th , H.G. (1969). Piaget and knowledge. Englewood C l i f f s , New Jersey: P r e n t i c e - H a l l , Inc. Gendreau, L . , Roach, T . , & Gendreau, P. (1973). Assessing the i n t e l l i g e n c e of aged persons: report on the Quick Test. Psychological Reports, 32, 475-480. 1 1 2 Glass , G.V, & Stanley, J . C . (1970). S t a t i s t i c a l methods in  education and psychology. Englewood C l i f f s , N . J . : Prentice H a l l Inc. , 410. Glenn, C.G. (1978). The role of episodic structure and of story length in c h i l d r e n ' s r e c a l l of simple s t o r i e s . Journal of  Verbal Learning and Verbal Behavior, 17, 229-247. Hartup, W.W. Sc Coats, B. (1972). Imi ta t ion : arguments for a developmental approach. In R.D. Parke (Ed.) Recent trends  in s o c i a l learning theory. New York: Academic Press, 63-75. Hays, W.L. (1973). S t a t i s t i c s for the s o c i a l sciences (2nd. E d . ) . New York: Hol t , Rinehart , and Winston, Inc. 594-611. H e r t e l , P.T. (1985). I so lat ion and adaptation in passage memory. Discourse Processes, 8_( 1 ) , 75-90. Honzik, M.P. (1984). L i f e span development. Annual Review of  Psychology, 35, 309-331. Horton, D.L. Sc M i l l s , C B . (1984). Human learning and memory. Annual Review of psychology, 35, 361-394. Houston, C , Sc Otto, W. (1968). Poor readers functioning on the WISC, Slosson In te l l igence Test and Quick Test. Journal of Educational Research, 62, 157-159. Howe, M.J . (1970). Repeated presentation and r e c a l l of meaningful prose. Journal of Educational Psychology, 61, 214-219. Hultsch, D.F. Sc Hickey, T. (1978). External v a l i d i t y of human development: the roet i ca l and methodological issues . Human  Development, 21, 76-91. Inhelder, B. , (1969). Memory and i n t e l l i g e n c e in the c h i l d . In D. E l k i n d , Sc ^ J . F l a v e l l (Eds.) Studies in Cognitive  Development. New York: Oxford Univers i ty Press. Inhelder, B. , Sc P iaget , J . (1964). The ear ly growth of log ic  in the c h i l d . New York: Harper Sc Row. Inhelder, B. , S i n c l a i r , H. , & Bovet, M. (1974). Learning and  the development of cogn i t ion . Cambridge, Mass. : Harvard Univers i ty Press . Jab lonsk i , E.M. (1974). Free r e c a l l in c h i l d r e n . Psychological B u l l e t i n , 8_1_, 522-539. 1 13 Jensen, A . R . , & Rohwer, W.D. (1963). The e f fec ts of verbal mediation on the learning and retent ion of paired associates by retarded adu l t s . Journal of Mental  Def ic iency , 68, 80-84. Joest ing , J . , & Joest ing , R. (1972). Ch i ld ren 's quick t e s t , p icture in te rp re ta t ion , and goodenough draw-a-person scores. Psychological Reports, 30, 941-942. Johnson, M.K., Bradsford, J .D , & Solomon, S .K. (1973). Memory for t a c i t impl icat ion of sentences. Journal of  Experimental Psychology, 98, 203-205. K a i l , R.V. (1979). The development of memory in c h i l d r e n . San Francisco: W.H. Freeman & Company. P. 18. K a i l , R.V. , J r . , & Hagen, J.W. (Eds.) (1977). Perspectives on  the development of memory and cogn i t ion . H i l l s d a l e , New Jersey: Lawrence Erlbaum Associates . Ker l inger , F .N. (1973). Foundations of behavioral research 2nd(Ed). New York: Ho l t , Reinhart & Winston, Chapter 6, 370. Ker l inger , F .N. (1979). Behavorial research: a conceptual  approach. New York: Ho l t , Rinehart & Winston. K in tsch , W. (1976). Memory for prose. In C.N. Cofer (Ed.) The  Structure of Human Memory. San Francisco: Freeman. K in tsch , W. (1974). The representation of meaning in memory. New Jersey: Lawrence Erlbaum Associates . K i r k , R.E. (1968). Experimental design: procedures for the  behavioral sc iences. Belmont, C a l i f o r n i a : Brooks/Cole publ ishing Company, 70-77. Klapp, S .T . , Marshburn, E. , & Lester , P.T. (1983). Short term memory does not involve the "working memory" of information processing: the dimise of a common assumption. Journal of  Experimental Psychology: General , June, 112, 245-264. Kobasigawa, A. (1977). Ret r ieva l s t rateg ies in the development of memory. In K a i l , R.V. , & Hagen, J.W. (Eds.) Perspectives on the development of memory and cogn i t ion . New Jersey: Lawrence Erlbaum Associates , 1977, 177-201. Ko lers , P.A. (1983). Perception and representat ion. Annual  Review of Psychology, 34, 129-166. 1 1 4 Kreutzer, M.A., Leonard, C , & F l a v e l l , J . H . (1975). An interview study of c h i l d r e n ' s knowledge about memory. Monographs of the Soc iety for Research in Ch i ld  Development, 40, 1-58. Kuhlman, E . S . , & Wolking, W.D. (1972). Development of w i t h i n -and cross-modal matching a b i l i t y in the auditory and v i sua l sense moda l i t ies . Developmental Psychology, 7, 365. Lange, G . , ' & Jackson, P. (1974). Personal organizat ion in c h i l d r e n ' s free r e c a l l . Ch i ld Development, 45, 1060-1067. Lesgold, A . M . , Lev in , J . R . , Shimron, J . , & Guttmann, J . (1975). P ic tures and young c h i l d r e n ' s learning from o ra l prose. Journal of Educational Psychology, 67, 636-642. Lev in , J . R . , Ghatala, E . S . , DeRose, T.M. , Wi lder , L . , & Norton, R.W. (1975). A further comparison of imagery and voca l i za t ion s t rateg ies in c h i l d r e n ' s d isc r iminat ion learn ing . Journal of Educational Psychology, 1975, 67, 141-145. Lev in , J . R . , McCabe, A . E . , & Bender, B.G. (1975). A note on imagery-indueing motor a c t i v i t y in young c h i l d r e n . Chi ld  Development, 46, 236-266. L ibb , W. J , , & Coleman, J . M . (1971). Corre lat ions between the WAIS and revised beta, wechsler memory scale and the quick test in a vocat ional r e h a b i l i t a t i o n center . Psychological  Reports, 29, 863-865. L iben, L .S . (1977). Memory in the context of cognit ive development: the Piaget ian approach. In R.V. K a i l , J r . , & J.W. hagen (Eds.) Perspectives on the development of memory  and cogn i t ion . New Jersey: Lawrence Erlbaum Associates , 297-231. Lof tus , E .F . , M i l l e r , D.G. & Burns, H .J . (1978). Semantic integrat ion of verbal information into v i s u a l memory. Journal of Experimental Psychology: Human Learning and  Memory, 4, 19-31. Lof tus , E .F . , & Palmer, J . C . (1974). Reconstruction of automobile dest ruc t ion : an example of the in teract ion between language and memory. Journal of Verbal Learning  and Verbal Behavior, 13, 585-589. Long, J . (1976). D i v i s ion of at tent ion between non-verbal s i g n a l s : a l l - o r - n o n e or shared processing? Quarterly  Journal of Experimental Psychology, 28, 47-69. 1 15 Maloney, M.P. , Steger, H.G., & Ward, M.P. (1973). The Quick Test as a measure of general i n t e l l i g e n c e in an urban community psych ia t r i c h o s p i t a l . Psychological Reports, 32, 823-827. Mandler, J .M . (1978). A code in the node: the use of story scheme in r e t r i e v a l . Discourse Processes, J _ , 14-35. Mandler, J . M . , & Johnson, N.S. (1977). Remembrance of things parsed: story structure and r e c a l l . Cognitive Psychology, 9, 111-151. Mandler, J . M . , & Murphy, C M . (1983). Subjective judgments of sc r ip t s t ructure . Journal of experimental Psychology:  Learning, Memory, and Cognit ion, 9(3 ) , 534-543. Marmurek, H . H . C , & Hol t , P.D. , & Coe, K. (1978). Presentation mode and r e p r e t i t i o n ef fects in free r e c a l l . American  Journal of Psychology, 9J_, 183-190. McGaugh, J . L . (1983). Hormonal inf luences on memory. Annual  Review of Psychology, 34, 297-323. Mednick, M.T. (1969). The v a l i d i t y of the Ammons quick test of i n t e l l i g e n c e . Psychological Reports, 24, 388-390. Meyer, B.F. (1975). The organization of prose and i t s e f fec ts  on memory. Amsterdam: North Holland Publ ishing Company. Pp. 71-76. Meyer, B.F. (1977). What i s remembered from prose: A function of passage s t ructure . In R.O. Freedle (Ed.) Discourse  production and comprehension. Norwood, New Jersey: Ablex Publ ishing Company, pp. 99; 307-336. M is t r y , J . J . , & Lange, G.W. (1985). Ch i ld ren 's organization of r e c a l l of information in scr ip ted nar rat i ves . Ch i ld  Development, 56(4), 953-969. Moely. B.E. (1977). Organization factors in the development of memory. In R.V. K a i l , J r . , & J.W. Hagen (Eds.) Perspectives on the development of memory and cogn i t ion . New Jersey: Lawrence Erlbaum Assoc iates , 203-206. Neisser , U. (1982). Memory observed: remembering in natural  contexts. San Francisco: W.H. Freeman and Company, 3-19. Nelson, K. (1981). Indiv idual d i f ferences in language development: impl icat ions for development and language. Developmental Psychology, 17, 170-187. 1 1 6 Nicholson, C L . (1977). Corre lat ions between the Quick Test and the Wechsler Inte l l igence Scale for c h i l d r e n - r e v i s e d . Psychological Reports, 40, 523-526. Northway, M.L. (1940). The concept of 'schema'. B r i t i s h  Journal of Psychology, 30, 316-325. Ornstein , P .A . , (1977). Naus, M . J . , & Stone, B.P. Rehearsal t r a i n i n g and development d i f ferences in memory. Developmental Psychology, 13, 15-24. Owen, S . V . , Froman, R.D. , & Moscow, H. (Eds.) (1981) Educational psychology. An introduct ion (2nd, ed.) Boston: L i t t l e , Brown and Company. P a r i s , S . C , & Lindauer, B.K. (1976). The role of inference in c h i l d r e n ' s comprehension and memory for sentences. Cognitive Psychology, 8, 217-227. P a r i s , S . C , & Lindauer, B.K. (1977). Constructive aspects of c h i l d r e n ' s comprehension and memory. In R.V. K a i l , J r . , & J.W. Hagen (Eds.) Perspectives on the development of memory  and cogn i t ion . H i l l s d a l e , New Jersey: Lawrence Erlbaum Associates , 35-59. P a r i s , S . C , & Upton,. L.R. (1976). Ch i ld ren 's memory for i n f e r e n t i a l re lat ionsh ips in prose. Ch i ld Development, 47, 660-668. Penman, K.A. , Christopher, J . R . , & Wood, G.S. (1977). Using gross motor a c t i v i t y to improve language ar ts concepts by t h i r d grade students. Research Quarter ly , 48, 134-137. Perlmutter , M. & Myers, N. (1979). Development of r e c a l l in 2- to 4- year -o ld ch i ld ren . Developmental Psychology, 15, 73-78. Pezdek, K., & M i c e l i , L. (1982). L i fe -span d i f fe rences in memory integrat ion as a function of processing t ime. Developmental Psychology, 18, 485-490. P iaget , J . (1969). The ch i lds conception of t ime. London: Routledge and Pau l . P iaget , J . (1976a). The c h i l d and r e a l i t y . New york: Penguin Books. P iaget , J . (1976b). The grasp of consciousness. Cambridge: Harvard Univers i ty Press. P iaget , J . (1962). P lay , dreams, and imi tat ion in chi ldhood. New York: Norton. 1 17 Piaget , J . (1967). Six psychological s tudies . New York: Random House. Piaget , J . , & Inhelder, B. (1973). Memory and i n t e l l i g e n c e . New York: Basic Books. Piaget , J . , & Inhelder, B. (1971). Mental imagery in the  c h i l d . (Translated by P.A. Chi l tonTi New York: Basic Books. Quay, L . C . , Hough, R.A. , Mathews, M. , & J a r r e t t , O.S. (1981). Predictors of communication encoding: age, socioeconomic s tatus , and cogni t ive a b i l i t y . Developmental Psychology, 17, 221-223. Rabinowitz, M. , & Mandler, J .M . (1983). Organization and information r e t r i e v a l . Journal of Experimental Psychology:  Learning, Memory, and Cognit ion, 9(3) , 430-439. Richardson, A. (1969). Mental Imagery. New York: Springer Publ ish ing Company, Inc . , 127-147. Rohwer, W.D., J r . , (1973). E laborat ion and learning in childhood and adolescence. In H.W. Reese (Ed.) Advances  in c h i l d development and behavior (Vol . 8 ) . New York: Academic Press. Rohwer, W.D., J r . , (1970). Imagery and contextual meaning. Psychological B u l l e t i n , 73, 404-414. Rohwer, W.D., J r . , & Dempster, F.N. (1977). Memory development and educational processes. In R.V. K a i l , J r . , & J.W. Hagen (Eds.) Perspectives on the development of memory and cogn i t i on . H i l l s d a l e , New Jersey: Lawrence Erlbaum Assoc iates , 407-436. Rohwer, W.D., J r . , & L i t rownik , J . (1983). Age and ind i v idua l d i f ferences in the learning of a memorization procedure. Journal of Educational Psychology, 75, 799-810. Rosner, S.R. (1971). The e f fec ts of rehearsal and chunking ins t ruc t ions on c h i l d r e n ' s m u l t i t r i a l free r e c a l l . Journal  of Experimental Ch i ld Psychology, 11, 93-105. Rubin, D.C. (1978a). A unit ana lys is of prose memory. Journal  of Verbal Learning and Verbal Behavior, 17, 599-620. Rubin, D.C. (1978b). A u x i l i a r y information f o r : a unit ana lys is of prose memory. Journal of Verbal Learning and  Verbal Behavior, _T7, 599-620. 118 Rumelhart, D.E. (1975). Notes on a schema for s t o r i e s . In D.G. Bobrow and A. C o l l i n s (Eds.) Representation and  understanding. New York: Academic Press, 211-236. Sachs, J . S . (1974). Memory in reading and l i s t e n i n g to d iscourse . Memory and Cognit ion, 2, 95-100. Scr ibner , S. (1984, January/Apr i l ) . Cognitive studies at work. The Quarterly Newsletter of the Laboratory of Comparative  Human Cognit ion, 6(1 & 2) , 1-46. S i e g e l , S. (1956). Nonparametric s t a t i s t i c s for the behavioral  sc iences. New York: McGraw-Hil l Book Company. Pp. 22-24. S i l ve rn S . B . , & Yawkey, T. (1977). Young c h i l d r e n ' s encoding of images based on motor and verbal modes of mediation. Psychology in the Schools, 14, 112-115. Spencer, N . J . (1973). Changes in representation and memory of prose. Unpublished Ph.D. d i s s e r t a t i o n , The Pennsylvania State Un ive rs i t y , September, (c i ted in Meyer, B.F. The  organizat ion of prose and i t s e f fec ts on memory. Amsterdam: North Holland Publ ish ing Company, 1975) Sp i ro , R . J . (1975). I n f e r e n t i a l reconstruct ion in memory for  connected discourse. Unpublished doctoral d i s s e r t a t i o n , Pennsylvania State Un ivers i t y , Un ivers i ty Park, Pennsylvania. (c i ted in D i S i b i o , M. Memory for connected d iscourse : a c o n s t r u c t i v i s t view. Review of Educational  Research, 1982, 52, 149-174). S t e i n , N . , & Glenn, C. (1975a). A developmental study of  c h i l d r e n ' s r e c a l l of story m a t e r i a l . St . Louis , M i s s o u r i : Washington Un ive rs i t y . Paper presented at the Society for Research in Ch i ld Development, Denver, Colorado. S t e i n , N. , & Glenn, C. (1975b). An analys is of story comprehension in elementary school c h i l d r e n : a test of schema. (ERIC Document Reproduction Service No. ED 121474) Strandberg, T . E . , G r i f f i t h , J . & Miner, L. (1969). Ch i ld language and screening i n t e l l i g e n c e . Journal of Communication Disorders , 2, 268-272. Thorndyke, P . E . , & Hayes-Roth, B. (1979). The use of schemata in the a c q u i s i t i o n and t ransfer of knowledge. Cognitive  Psychology, J M , 82-106. Thorndyke, P.E. (1977). Cognitive structures in comprehension and memory of narrat ive d iscourse. Cognitive Psychology, 1, 77-110. 1 19 Trabasso, T. , & R i l e y , C A . (1973). An information processing  analys is of t r a n s i t i v e inferences. Princeton Un ivers i t y . (ERIC Document Reproduction Service No. ED 091043) T rav is , L.D. (1984). A question of substance: the metaphor  problem and teacher as content for s e l e c t i o n . (paper presented to the education department seminar at Edinburgh Un ivers i t y , A p r i l ) The Univers i ty of B r i t i s h Columbia, Vancouver, B.C. T rav is , L .D . , ( in press) . Truths, concerns and consequences: on psychology's place in and or ientat ion to education. Interchange. T rav is , L .D . , & White, W.B. (1979). Experimental manipulation of the r e c a l l of narrat ive mater ial by f i v e - y e a r - o l d s . The  Alberta Journal of Educational Research, 25, 137-146. Tulv ing, E. , & Thompson, D.M. (1973). Encoding s p e c i f i c i t y and r e t r i e v a l processes in episodic memory. Psychological  Review, 80, 352-373. V i o l a t o , C , White, W.B., & Trav is , L.D. (1984). Some concurrent c r i t e r i o n - r e l a t e d V a l i d i t y data for the quick Test based on three Canadian samples. Psychological  Reports, 54, 775-782. Vygotsky, L . S . , (1978). Mind in soc iety : the development of  higher psychological processes. M. Cole, V. John-Steiner , S. Scr ibner , & E. Souberman (Eds. ) , Cambridge, Mass. : Harvard Univers i ty Press ( o r i g i n a l l y publ ished, 1930), 19-56; 79-104. Wechsler, D. (1958). The measurement and appra isa l of adult  i n t e l l i g e n c e , (4th .Ed . ) . Bait imore, Md. : Wil l iams & W i l k i n s . White, W.B. (1978). Recal l of narrat ive mater ia l in f ive year  o lds . Unpublished masters t h e s i s , The Univers i ty of B r i t i s h Columbia, Vancouver, B.C. Wickelgren, W.A. (1981). Human learning and memory. Annual  Review of Psychology, 32, 21-52. Winer, B . J . (1962). S t a t i s t i c a l p r i n c i p a l s in experimental  design. New York: McGraw H i l l Book Company Wingf ie ld , A. (1979). Human learning and memory: an in t roduct ion . New York: Harper and Row. 120 Yussen, S . R . , Gagne, E . , G a r r i u l o , R., & Kunen, S. (1974). The d i s t i n c t i o n between percievinq and memorizing in elementary  school . Un ivers i t y ol Wisconsin. (ERIC Document Reproduction Service No. ED 098538) 121 APPENDIX A Rationale for the Narrat ive Content The nar rat ive used in th i s study i s an o r ig iona l composition. The use of same ensured-that no ch i ldren had seen or heard the story pr io r to i t s presentat ion. In t h i s way one gains experimental contro l by ensuring a l l subjects have an i d e n t i c a l amount of experience with the mater ial on which the i r r e c a l l i s tes ted . Within the story there are 7 instances of non-motoric act ion sequences that can not be overt ly acted out. Of these 7, only 2 are complete sentences. One of these sentences deals with pure thought (and no a c t i o n ) ; and the other sentence deals with thought that contains ac t ion . In Rumelhart's (1975) story categor i za t ion , both of these sequences (as wel l as others) would be categorized as ' i n t e r n a l knowledge' and are accounted for in the measuring instrument proposed. Moreover, previous studies (Travis & White, 1979; White, 1978) have contained s imi la r categories in the i r story s t ructure . Upon invest igat ion of the resu l ts from these s tud ies , i t has been found that some chi ldren from a l l treatment condit ions (s imi lar treatments to those proposed for th i s study) r e c a l l e d some of these story 'categor ies ' during free r e c a l l (as wel l as the other ' c a t e g o r i e s ' ) . Moreover, p i l o t studies have been c a r r i e d out with ch i ldren in a l l age-span groups and treatment cond i t ions , and some of these instances of information (as wel l as the other instances of information defined by the categories) have been reca l led by some ch i ldren in a l l the age-spans and treatment condi t ions . Therefore, since a l l categories or instances of information appeared to be reca l led regardless of treatment or age-span, there was no reason to suppose that any c h i l d in any age-span or treatment condit ion would not be able to r e c a l l these instances of information. 122 APPENDIX B Rationale for Di rect ions for E l i c i t a t i o n of Free Recal l The present study was designed to invest igate how age and encoding condit ions interact to influence the extent to which ch i ldren could produce from memory complete and accurate reconstructions of the narrat ive content which they encountered. However, the nature of the d i rec t ions or request for the reconstruction can influence performance. Since between age (within condit ion) and between condit ion (within age) comparisons were made, one had to ensure that (1) a l l respondents would interpret the task requirements (for r e c a l l ) in the same manner; and (2) that a l l respondents, without prejudice to condit ion or age, would perform as wel l as they were ab le . In order to develop the ins t ruct ions which f u l l f i l l e d the above c r i t e r i a , p i l o t studies were car r ied out which tested subjects (N=36) in a l l age-spans for r e c a l l and interpretat ions of the various requests. Some chi ldren (22%) who were given the d i rec t ions to r e c a l l in the 'exact ' context refused to give any account of the story with the usual . response being, "I can' t t e l l i t back ' e x a c t l y ' " , when questioned. When probed or cued to r e t e l l parts of the story , these ch i ldren (80%) as well as others who stated that they ' fo rgot ' some parts (60%) were able to r e c a l l some parts of the s tory . Of those subjects who were given the same d i rect ions as in th i s study, about 5% refused to r e c a l l and about 5% stated they ' fo rgot ' some par ts . Moreover, when questioned, almost a l l (95%) of the ch i ldren given the ins t ruct ions said that they interpreted the ins t ruc t ions to mean that they were to r e t e l l the story just as they heard i t . Accordingly , requests for exact reproduction can be expected to i n f l a t e fa lse negatives because some chi ldren refuse to respond or say less than they know when given what they seem to percieve as an impossible task. On the other hand, requests which do not indicate that the most complete and accurate possible reconstruction i s desired can be expected to camouflage the magnitude of d i f ferences in r e c a l l . In such circumstances, some ch i ldren might not interpret the task requirements to mean that they should reconstruct the story as accurately and completely as they can while others may assume otherwise. However, on the basis of the p i l o t study evidence where the experimental inst ruct ions ( i . e . " r e c a l l as best as you can.") indicated that an exact reconstruct ion was not mandatory, but that the i r best e f f o r t s in reconstruct ing the story was desi red, greater uniformity in in terpretat ion of task requirements and greater uniformity in readiness to expend the i r best e f f o r t s was 1 2 3 expected. This set of ins t ruc t ions then, seemed to e l i c i t more responsiveness than requests for exact reproduction and thus fa lse negatives were minimized while ch i ldren could be expected to s t r i v e for complete and accurate reconstruct ions. Under these condit ions the magnitude of rea l d i f ferences should not be camouflaged as d i f ferences might be i f ch i ldren were not given to understand that the i r best e f f o r t s were being s o l i c i t e d . On the other s ide , the adoption of standard protocols ( i . e . the word unit and category measures) which defined the l i m i t s within which deviat ions (transformations of verbatim text) were acceptable, cont ro l led the p o s s i b i l i t i e s for fa lse p o s i t i v e s . 124 APPENDIX C Questions Asked about the Story (followed by acceptable answers) 1 2 3 8: 9 : 1 0 : 11 12 13 14 15! 16: 17: 18 1 9 : 20 21 22 23 24 What animal was the story about? A. cougar, ca t . What was the cougar's name? A. Rufus. Where did he l i v e ? A. In a hole , cave, den, tunnel/ underground. When did he leave h is home to look for food? A. In the (early) morning. What did he do as he went down the t r a i l ? A. He turned his head from side to side (and/or) to look for food. The longer he walked along the forest t r a i l , what d id he begin to do? A. Walk faster / run. As Rufus was walking down the forest t r a i l , what did he see when the forest disappeared? A. A town / houses. What side of the t r a i l was the town on? A. l e f t ( s ide ) . What did Rufus see a l l around him in the town? A. Houses. How did Rufus fee l when he looked at a l l the houses around him in the town? A. Dizzy and/or hungry (one point each). What did Rufus see in the town that he thought he could eat? A. A ju icy bone / a bone / food. What happened when he dived at the bowl? A. The bowl c la t te red / i t c l a t t e r e d . Who woke up with a l l the noise? A. A dog. What did Rufus do when the dog barked? A. His ears stood up / he began to creep away. What happened when Rufus was creeping away? A. Something made a loud noise next to him. When he heard a loud noise next to him, what d id he do? A. (Jumped) high into the a i r . After he jumped high into the a i r , what did he do? A. Ran away / ran home. How did Rufus run? A. As fast as h is legs could carry him / as fast as he could / fast / q u i c k l y . What d id Rufus have for breakfast? A. Nothing but food for thought / nothing. Did you l i k e the story? A. Omit. Would you l i k e to hear more s to r ies l i k e t h i s ? A. Omit. Would you l i k e to read s tor ies l i k e th i s? A. Omit. Do you l i k e s to r ies about animals? A. Omit. Do you l i k e s to r ies about people? A. Omit. APPENDIX D Word Units from Story 1) early 2) morning 3) black 4) cougar 5) named 6) rufus 7) l e f t 8) hole 9) home 10) walked 1 1 ) forest 12) t r a i 1 13) looking 14) something 15) eat 16) turned 17) head 18) side 19) other 20) longer 21 ) faster 22) ran 23) l i t t l e 24) s t i l l 25) searched 26) suddenly 27) disappeared 28) l e f t 29) saw 30) town 31 ) thought 32) food 33) leaped 34) soon 35) nothing 36) houses 37) round 38) nearly 39) f e l l 40) over 41 ) f e l t 42) dizzy 43) hungry 44) sight 45) bowl 46) ju icy 47) bone 48) made 49) forget (cat / animal) (he / i t / cat) (went/leave/came out/jumped out/got out) (cave) (walking) (woods / bushes / jungle) (path / road) (looked / f inding) (some) (food / bone) (twisted / swing / spinned / twir led) (around / back / for th / both ways) (around) (more) (fast / quicker) (run / scampered) (some / a b i t ) (sudden) (gone / no more) (d i rect ion - l e f t hand) ( c i t y / v i l l a g e ) ( to ld himself) (went / jumped / hopped / scampered) (no) (around) (almost) (down) (got) (saw) (dish / p late ) (forgot) 126 50) good 51 ) meal 52) dived 53) c l a t t e r 54) awoke 55) dog 56) bark 57) ears 58) stand 59) up 60) comming 61 ) c loser 62) pressed 63) body 64) ground 65) began 66) creep 67) loud 68) noi se 69) next 70) jumped 71 ) high 72) a i r 73) legs 74) carry 75) breakfast 76) food for thought (dinner) (woke) (barking / growl) (stood / perked) (came / got) (nearer) (flattened/crouched/pushed/laid/lay f l a t ) (himself) (started) (sneaking / crawl) (big) (bang / sound) (beside) (leap) (up) (take) (food) 127 APPENDIX E Categories 1) One early morning, 2) a black cougar named Rufus 3) l e f t his hole that was home, 4) and walked along a forest t r a i l . 5) He was looking 6) for something to eat. 7) as he looked 8) he turned h is head from one side to the other . 9) The longer he looked 10) the faster he walked. 11) He even ran a l i t t l e , 12) with his head s t i l l turning 13) as he searched. 14) Suddenly the forest disappeared 15) and to h is l e f t 16) he saw a town. 17) He thought, "There must be food there . " 18) He leaped 19) toward the town.. 20) Soon he could see nothing but houses 21) when he turned round and round. 22) As he did so he nearly f e l l over. 23) He f e l t dizzy - and hungry. 24) The sight of a bowl with a ju icy bone in i t 25) made him forget he was dizzy and hungry. 26) "That w i l l make a good meal ." He s a i d . 27) Rufus dived 28) at the bowl. 29) The c l a t t e r awoke a dog 30) whose bark 31) made the cougar's ears stand up. 32) The dog's bark was coming c loser and c l o s e r . 33) Rufus pressed h is body close to the ground 34) and began to creep away. 35) But suddenly something made a loud noise next to him. 36) Rufus jumped high into the a i r 37) and ran home 38) as fast as his legs could carry him. 39) He had nothing to eat for breakfast 40) but food for thought: 41) A bone in a bowl puts a cat in the hole . 128 APPENDIX F Figure 2B Mean r e c a l l performance (PT-2) on the cued r e c a l l measure by each treatment condi t ion group (EC, IC, SC, and SIRC condit ions for each age-span group 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0096785/manifest

Comment

Related Items