UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Fast mapping and success in French immersion programs Houston, Ruth Anne 1990

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

Item Metadata

Download

Media
831-UBC_1990_A6_7 H68.pdf [ 2.57MB ]
Metadata
JSON: 831-1.0098089.json
JSON-LD: 831-1.0098089-ld.json
RDF/XML (Pretty): 831-1.0098089-rdf.xml
RDF/JSON: 831-1.0098089-rdf.json
Turtle: 831-1.0098089-turtle.txt
N-Triples: 831-1.0098089-rdf-ntriples.txt
Original Record: 831-1.0098089-source.json
Full Text
831-1.0098089-fulltext.txt
Citation
831-1.0098089.ris

Full Text

FAST MAPPING AND SUCCESS IN FRENCH IMMERSION PROGRAMS By RUTH ANNE HOUSTON BA. , University of Western Ontario, 1988 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES The School of Audiology and Speech Sciences We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA August 1990 © Ruth Anne Houston, 1990 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. The University of British Columl Vancouver, Canada Department DE-6 (2/88) ii A B S T R A C T A s a result of the high price paid in t ime and concern by s tudents, parents and educators in educat ing a chi ld in F rench Immersion only to find that s/he would be better off in an Engl ish c l a s s r o o m , a predictor is needed to dec ide early in a chi ld 's life whether or not s/he is a good candidate for F rench Immersion. A 'good candidate ' wou ld be s o m e o n e w h o wou ld be able to learn F rench and , as a corol lary, not be hand icapped academica l l y by be ing instructed in French. The present study is an examinat ion of the L2 learning aspec t of F rench Immersion. In part icular this paper will explore the possibi l i ty that 'fast mapping' . the ability to quickly make a partial representat ion of the mean ing , form and use of a word after hearing it only a few t imes, may be a predictor of s u c c e s s in acquir ing a s e c o n d language irrespect ive of overal l a c a d e m i c ach ievement . In a procedure adapted from Dol laghan (1985) e ighteen s tudents in G r a d e 2 French Immersion, seven teen in G r a d e 3 French Immersion and five former F rench Immersion students now in the G r a d e 3 Eng l ish program were e x p o s e d to an unusual ly s h a p e d , as yet unnamed object in the course of a hiding g a m e . Th is object w a s randomly ass i gned one of a set of nonsense names . The chi ldren were then admin is tered a 10 minute oral F rench C o m p r e h e n s i o n Tes t as a distractor before be ing tested for their comprehens ion and product ion of the new word . T h e sco res on these tasks, wh ich are an indicat ion of "fast mapp ing" ski l l , did not correlate with i) number of years exposure to . a s e c o n d language , ii) age , iii) teacher ratings of oral F rench or iv) a c a d e m i c ability, v) oral French comprehens ion or vi) inclusion in F rench Immersion. Th is sugges ts that this set of fast mapping tasks is not a good predictor of s u c c e s s in s e c o n d language learning or F rench Immersion. More research is needed to ascerta in the reciprocal effect of L2 learning on fast mapping ski l ls, the deve lopment of fast mapping ski l ls with age , and the effect of a more complex fast mapping task on the fast mapp ing per formance of schoo l age chi ldren. T A B L E O F C O N T E N T S Abst ract Tab le of Conten ts List of T a b l e s A c k n o w l e d g e m e n t s Chap te r 1 Introduction 1.0 Introduction 1.1 Lex ica l Learn ing 1.2 Fas t Mapp ing 1.21 S u m m a r y 1.3 Fas t Mapp ing and Language Acquis i t ion 1.31 S e c o n d Language Learn ing 1.32 S e c o n d Language Instruction 1.33 S u m m a r y 1.4 S u c c e s s in F rench Immersion 1.41 Predictors of S u c c e s s 1.42 S u m m a r y 1.5 Th is Study Chap te r 2 Me thod 2.1 Overv iew 2.2 Sub jec ts 2.3 Procedure 2.31 Exposure T a s k 2.32 F rench C o m p r e h e n s i o n Tes t iv 2.33 C o m p r e h e n s i o n T a s k 26 2.34 Product ion T a s k 26 2.35 Recogni t ion T a s k 26 2.36 Li teracy Tes ts 26 2.37 T e a c h e r Rat ings 27 Chap te r 3 Resu l t s 3.1 Fas t Mapp ing 28 3.11 Fas t Mapp ing and G r o u p Membersh ip 28 3.12 Fas t Mapp ing and Ora l F rench Product ion 29 3.13 Fas t Mapp ing and G e n e r a l A c a d e m i c Abil ity 30 3.14 Fas t Mapp ing and W o r d Recogni t ion 31 3.15 Fas t Mapp ing and Word Mean ing 32 3.16 Fas t Mapp ing and Ora l F rench C o m p r e h e n s i o n 32 3.17 Fas t Mapp ing and A g e 32 3.2 Re la t ionsh ips Be tween M e a s u r e s of F rench Prof ic iency 33 Chap te r 4 D i scuss ion 4.1 Fas t Mapp ing 35 4.11 S u m m a r y 37 4.2 T h e Prob lem of Predict ion 38 4.21 S u m m a r y 39 R e f e r e n c e s 40 Append ix A : T e a c h e r Quest ionna i re 42 LIST OF TABLES Table I. Percentage correct in various fast mapping studies. Table II. List of target words and their distractors. Table III. Distribution of fast mapping production scores by instructional group. Table IV. Distribution of fast mapping production scores by teacher rating of oral French production skill. Table V. Distribution of fast mapping production scores by teacher rating of math abilities. Table VI. Distribution of fast mapping production scores by age. Table VII. Spearman's rank correlation coefficients for relationships between various measures of French proficiency. VI ACKNOWLEDGEMENTS Sincere thanks to the students, parents, teachers and staff of the Grades 2 and 3 classes who supported and participated in this study. 1 CHAPTER 1 INTRODUCTION 1.0 INTRODUCTION As a result of the high price paid in time and concern by students, parents and educators in educating a child in French Immersion only to find that s/he would be better off in an English classroom, a predictor is needed to decide early in a child's life whether or not s/he is a good candidate for French Immersion. A 'good candidate' would be someone who would be able to learn French and , as a corollary, not be handicapped academically by being instructed in French. The present study is an examination of the L2 learning aspect of French Immersion. In particular this paper will explore the possibility that fast mapping, the ability to quickly make a partial representation of the meaning, form and use of a word after hearing it only a few times, may be a predictor of success in acquiring a second language irrespective of overall academic achievement. This chapter will present a review of the current literature concerning fast mapping and predictors of success in French Immersion. 1.1 LEXICAL LEARNING A fascinating question in the study of language acquisition is: How do children analyze the complex, continuous string of sounds that comprise speech? This question encompasses i) How do they determine the relevant units for speech production/perception? (phones) and ii) How do they determine the relevant units for language production/comprehension in : a) phonemes, b) words (morphology and semantics), c) sentences (morphology, semantics and syntax) and d) discourse (morphology, semantics, syntax and pragmatics). The present project will address only the problem of the comprehension and production of words. In order to be able to understand and produce a word correctly a child must be exposed to 2 a production of that word from another speaker. This production of the word, from which the child will build a representation, will be embedded in several different but related contexts. First, the word will often be part of a continuous stream of sounds which somehow must be broken down perceptually into meaningful units. Children seem to do this based on which units they have heard before and paid attention to (Peters, 1983). Some children extract units the size of adult phrases which include several morphemes (e.g. 'What's that?'). Others extract small ones the size of adult words (e.g. 'doggie') but all normal children divide the speech stream into manageable 'chunks' which I will refer to as words. Second, the word is embedded in a phrase or clause, which in turn is a product of the syntactic system of the language. For adults there is no one-to-one correspondence between word form class (e.g. noun) and meaning (e.g. object). When simple morphological rules are applied, words can change syntactic form while retaining the essence of their meaning (e.g. 'burglar' vs. 'burglarized'). The language addressed to, and used by, young children is simpler than this. Syntax is apt to be a more reliable cue to meaning (e.g. 'the' is predictably followed by an object label). For a child learning language a partial understanding of syntax can give clues as to whether a word refers to the name of an object (noun), an attribute of objects (adjective) or an action objects perform (verb) ( Maratsos and Chalkley, 1980; Peters, 1983). In this way the child will be able to, consciously or unconsciously, decide things like: "That word ended in 'ing' so it must be an action", and thus be able to narrow down the possible field of meanings for the new word. Third, the word will be embedded in the structure of the discourse going on between the child and the conversational partner. While young children have difficulty producing the complex cohesive ties adults use to delineate old versus new information (Ruthven, 1989), they show an early ability to identify the topic of a conversation (Carroll, 1987) and to stick to that topic. Thus, the child will be able to predict what the speaker will mean next, irrespective of the form that meaning comes in 3 s imply by be ing famil iar with the topic of conversat ion. Final ly, the first time a chi ld hears a new word will be a point in t ime when the child is in a part icular p lace , paying attention to certain activit ies or objects in the envi ronment. This n o n -l inguistic context will a lso provide c lues as to what the word m e a n s and when it shou ld be used . The chi ld may overgenera l ize or undergenera l ize use of the word on the bas i s of ana lys is of this context, but will have a genera l s e n s e of what 's 'right'. T h u s , in order to unders tand a new word the chi ld must be ab le to recogn ize units of the language , have a genera l idea of which syntact ic forms cor respond to wh ich seman t i c content in that language , and der ive c u e s from the d iscourse and non- l inguis t ic contexts a s to what the word may m e a n . In order to produce the word in an adul t - l ike way the chi ld must have done the above to s o m e extent and must a lso interpret the present l inguistic and non- l ingu is t ic contexts a s s o m e h o w be ing simi lar to that in wh ich s /he first heard the word . S /he must then be able to retrieve the appropr iate phono log ica l information from memory in order to produce it. 1.2 F A S T M A P P I N G Chi ld ren learn language in an amaz ing ly short per iod of t ime s o it shou ld not be surpr is ing that they are, speci f ical ly , very good at learning new words. Templ in (cited in R ice and W o o d s m a l l , 1988) reported that chi ldren between the ages of 1;6 and 6;0 learn an average of nine new words per day. Th is ability to quickly acquire information about a new word from both the l inguistic and the non- l ingu is t ic context is known as "fast mapp ing" (Carey , 1978). C a r e y and Bartlett (1978) hypothes ized that fast mapping is the first s tage in the acquisi t ion of a new lexical i tem, the result of wh ich is a partial representat ion in the chi ld 's memory of the word 's phono log ica l character is t ics, its mean ing , the l inguistic and non- l ingu is t ic context in which it occur red or any combinat ion of the above . The s e c o n d , "ex tended" , s tage of acquisi t ion involves 4 the s low ref inement of this fast mapped representat ion into the adult form of the word . Th is second s tage may take years to complete. Th is hypothes is prov ides one explanat ion for the observat ion that smal l chi ldren may use the word 'kitty' correctly to refer to their pet but a lso incorrectly to refer to any sma l l , furry an ima l . Ye t there are no normal adults who call all sma l l furry an ima ls 'kitty'. Our understanding of the full mean ing and connotat ion of a word grows with our exposure to and interaction with language. S tud ies into the first s tage of lexical acquis i t ion, the fast mapp ing s tage, have taken a variety of forms. C a r e y and Bartlett (1978) exp lored 3 and 4 - y e a r - o l d chi ldren's fast mapping of the new colour term chromium (olive) in a natural c l ass room situation. The chi ldren were individually told " Br ing me the chromium tray, not the blue one , the chromium one" when p resen ted with two trays differing only in colour. O n e w e e k later the chi ldren were tested to s e e if they unders tood the word, if they knew that it w a s a co lour term and if they could produce it. At least half of the chi ldren were able to demonst ra te that they had learned someth ing about the word after being e x p o s e d to it only o n c e , one w e e k previously. In a s tudy of 35 2 through 5 - y e a r - o l d s Dol laghan (1985) e x p o s e d chi ldren to a novel , unc lass i f iab le object ca l led a koob in the course of a hiding game with a puppet. After hiding s o m e famil iar objects e a c h chi ld w a s a s k e d to hide the koob (the remaining object) in the remaining hiding p lace . T h e y were then immediate ly tested to s e e if they unders tood the word ("Give me the koob"), cou ld produce it ("What 's this?"), could recogn ize it if they couldn't p roduce it ("Is this a koob, a soob or a teed?") and if they could remember s o m e non- l inguis t ic information concern ing the object ( "Where did you hide this?"). By age 4 all the chi ldren in the study could perform the initial task without ask ing for clarif ication. A n average of 81% of the chi ldren could perform the comprehens ion task and 45% could produce two of the three p h o n e m e s in the word (at this point they had heard the word twice). Of those who did not attempt to name the object, 62% cou ld recogn ize the word when 5 they heard it again. Dollaghan interpreted these results to mean that phonetic information, as opposed to semantic or syntactic information, is the most vulnerable aspect of fast mapping, especially the retrieval of this phonetic1 information for spontaneous production of the new word. While not designed to look for developmental differences in fast mapping, the response percentages at each age level revealed some interesting trends in performance across ages. General trends seem to be i) an increasing ability to perform the initial task with age, ii) a constant ability to comprehend the new word from ages 2-5 years and iii) an extremely variable ability to produce a new word ( 2 yrs.- 57%, 3 y r s - 0%, 4 y r s - 45%, 5 yrs - 71% ). Further research has substantiated the claim that children are able to remember a significant amount of information as long as one week after the initial exposure task. The ability to understand a new word in running speech seems constant from age 2 to 5, while the ability to produce a new word varies greatly. This variation in production ability is not, apparently, because of increasing age. Perhaps elements of personality, such as risk taking, and personal learning style are responsible for the different productive abilities at one age. The two studies mentioned above epitomize the simplest type of fast mapping task. Each child is expected to learn one new word either real or 'nonsense' ( and all unfamiliar words are, in a sense, initially nonsense ). The word in question refers to either an object or a salient characteristic of objects. The child manually experiences the object in question and there is little ambiguity as to which object/characteristic matches the new word. Other researchers have investigated the effect of a variety of factors on fast mapping ability by manipulating the basic task paradigm. Factors studied include the effect of i) form class types other than nouns and adjectives, ii) content class types other than object and characteristic, iii) multisyllabic vs monosyllabic targets, iv) degree of interaction with the object the word represents, v) multiple targets presented in multiple :It is unclear whether Dollaghan means phonetic or phonological information in this instance. 6 contexts and vi) age di f ferences among the subjects. He ibeck and Markman (1987) invest igated fast mapping of unfamil iar words from the semant i c doma ins of colour, s h a p e and texture in 2 to 4 - y e a r - o l d s us ing C a r e y and Bartlett 's (1978) contrast ive exposure procedure ("Bring me the X one , not the Y one , the X one") . After a 10 minute conversat iona l de lay the chi ldren were a s k e d to descr ibe the one target colour/shape/ texture term they had been e x p o s e d to from an array of severa l famil iar (objects and atr ibutes they could name) and unfamil iar ( objects and attributes for wh ich they had no names) objects. To further probe the chi ldren 's understanding of the doma in of the new word , the chi ldren were a s k e d hyponym quest ions of the type: " S e e this box? It's not f ibrous [target] b e c a u s e it's " . If the chi ld responded with a texture term it w a s inferred that they knew "f ibrous" to be a texture term. Final ly the chi ldren were a s k e d to identify the target adject ive from an array of distractors, showing that they understood the new word (e.g. " C a n you s h o w me the char t reuse one?"). T h e results s h o w e d that fast mapping occur red in all three doma ins al though less information w a s retained for texture terms than for colour, and less for co lour than for s h a p e . There w a s no signif icant age effect for the comprehens ion task (average score for all three age groups w a s 61%). T h e ability to perform the hyponym task did inc rease with age (2 y r s - 64%, 3 y r s . - 89%, 4 y r s . - 95%) but this w a s not true of the product ion task ( 2 yrs - 13%, 3 yrs - 22%, 4 y r s . - 19%). A subsequen t study presented in the s a m e article found no signif icant di f ference in fast mapp ing between si tuat ions in which the target word w a s presented with explicit l inguistic contrast as in the first exper iment , and when it w a s presented with an implicit l inguistic contrast ( e . g . "Bring me the X one , not the other one") . They conc luded from this tha t " w h e n the context is compel l ing, chi ldren do not s e e m to rely on or benefit f rom hear ing a new word contrasted with a famil iar word" (He ibeck and M a r k m a n , 1987:1030) but are able to rely on the non- l ingu is t ic context. He ibeck and M a r k m a n cite a study in progress which shows that chi ldren rely on non- l ingu is t ic information more 7 than l inguist ic information and will interpret the most unusua l and sal ient shape/colour/ texture present as be ing the referent of the new word (He ibeck and M a r k m a n , 1987). He ibeck and Markman a lso found smal l but signif icant correlat ions be tween prior vocabulary within the seman t i c doma ins they studied (i.e. colour, s h a p e or texture ) and fast mapping per formance with a new word from that doma in . Chi ldren knew more s h a p e terms than colour terms and more colour terms than texture terms. They sugges t that chi ldren may fast map better with words from a part icular doma in either b e c a u s e the chi ldren have learned more words in that domain or b e c a u s e they find the domain to be more sal ient than others. They point out, however , that these smal l correlat ions be tween vocabu lary and fast mapp ing cannot account for the whole effect of doma in dom inance . Ca tegory labels have been found to have spec ia l status for chi ldren ( M a c n a m a r a , 1982). It is poss ib le that the s h a p e terms may have been interpreted as nouns rather than adject ives. S i n c e nouns s e e m eas ie r for chi ldren to learn this cou ld explain the dif ference in s c o r e s be tween the semant i c types of words u s e d in this exper iment. In order to invest igate this possibi l i ty He ibeck and Markman compared the fast mapping per fo rmance of chi ldren exposed to s e n t e n c e s like "Br ing me the hexagon , not the tr iangle" and those e x p o s e d to "Bring me the hexagona l one, not the tr iangular one" . There w a s no signif icant di f ference in fast mapping per fo rmance be tween the two groups, suggest ing that the di f ference in s c o r e s between the shape and colour/texture groups w a s not mere ly the result of chi ldren interpreting s h a p e terms as nouns. In a presentat ion at the Amer i can S p e e c h - L a n g u a g e - H e a r i n g Assoc ia t ion Annua l Conven t ion in 1987, Ape l and Kamh i d i s c u s s e d their exper iments involving both normal and language impaired preschoo lers . O n e exper iment (Apel and K a m h i , 1987) w a s conducted to d i scove r the effect of mult isyl labic words on fast mapp ing in language impaired chi ldren. They found that there w a s no signif icant di f ference in comprehens ion and recognit ion s c o r e s be tween children e x p o s e d to monosy l lab ic words and chi ldren e x p o s e d to mult isyl labic words . None of the language 8 impaired chi ldren were able to produce any of the words. The other exper iment , involving normal language chi ldren, found a tendency for product ion sco res to be lower after a half hour pause than immediate ly after the exposure task, even though the comprehens ion s c o r e s were the s a m e in both condi t ions. Fo r both normal and language impaired chi ldren, phonolog ica l complexi ty w a s not found to affect fast mapping signif icantly. R ice and W o o d s m a l l (1988) expanded the s c o p e of fast mapp ing studies by investigating fast mapp ing of object, act ion, attribute and affective state words presented to 3 and 5 - y e a r - o l d s in two 6 minute televis ion programs. The chi ldren d id not interact in any w a y with the programs. The words were unfamil iar to p reschoo le rs (e.g. g ramophone , mal ic ious) and e a c h story conta ined 20 new words , each appear ing five t imes in a variety of contexts. They found that the 5 - y e a r - o l d s learned more than the 3 - y e a r - o l d s even when prior vocabulary w a s taken into account . W o r d s which proved e a s y to remember and use were those from the form c l a s s e s wh ich occur ear ly in language deve lopment . W o r d types which are acqu i red first in language acquis i t ion, such a s object, action and attribute words , were fast m a p p e d relatively easi ly , whi le affective state wo rds , wh ich are acquired later in language learning, were relatively difficult for all the chi ldren. In contrast to the previously ment ioned s tud ies R ice and Woodsma l l p resented the words in a variety of semant ic , grammat ica l and env i ronmenta l contexts, often with more than one new word in a sen tence . With words p resen ted in this variety of contexts the 5 - y e a r - o l d s unders tood an average of 4.87 new words and the 3 - y e a r - o l d s 1.56 new words from watching 15 minutes of careful ly eng ineered televis ion. No product ion da ta were co l lected so we cannot know whether this trend of increas ing sco res with age wou ld have occur red in that task a s wel l as the comprehens ion task. D ick inson (1984) compared the fast mapping of schoo l age chi ldren in three condit ions. N o u n s and adject ives were presented in conversat iona l contrast as in C a r e y and Bartlett (1978), in a story or as a definit ion. In testing the chi ldren were a s k e d to dec ide whether each of a list ( including 9 the target) of famil iar and nonsense words were 'words' . Al l of the chi ldren, regard less of condit ion, recogn ised the target as being a 'word' after only one exposure . Then the chi ldren were ask ed to dec ide if s o m e sen tences used the 'words' correctly. In the first exper iment (when there was no p a u s e be tween the exposure and testing tasks) only 25% of first g raders were able to do this correct ly, compa red to 78% of sixth graders . In the s e c o n d exper iment (when there w a s a pause of three to s e v e n days be tween exposure and testing), there were no signif icant d i f ferences within the age groups inc luded in per formance of this task ( 4 - 5 yrs - 21%, 6-7 yrs - 28 %, 8-9 yrs - 44%). The s e c o n d study d id not include G r a d e 6 students. T h e only product ion da ta col lected were the chi ldren 's definit ions of the target words in the story and definition condi t ions. Accep tab le definit ions were s y n o n y m s and sen tences correct ly using the new word . 2 First g raders did equal ly poorly at this in both the story (0%) and definition condi t ions (3%) condit ions. Sixth g raders did much better (60%, 67%), p resumab ly b e c a u s e of their super ior metal inguist ic knowledge. There w a s no difference in per formance on this product ion test between tests conducted immediate ly after exposure and those conduc ted three to s e v e n days afterwards. There w a s a tendency for new words referring to per iphera l , not easi ly categor izab le objects/colours to be eas ie r to remember . Th is appeared to be b e c a u s e of an overal l " res is tance to redundancy" (Dick inson 1984:371) in the conceptua l sys tem. 1.21 S u m m a r y T h u s , the data on the fast mapping phenomenon to date indicate that it occurs in chi ldren as young as 2 years o ld and may be respons ib le for the 'word burst' p h e n o m e n o n wh ich al lows children to learn a s many as nine new words a day between the ages of 1;6 and 6;0. Y o u n g chi ldren are apparent ly ab le to form partial representat ions in memory of words presented with or without explicit 2No data is given concerning what percentage of the children correctly produced the word itself, or about the precise distribution of the conditions. 10 l inguistic contrast and with or without interaction with objects . They can map words from the l inguistic doma ins of real or nonsense object, act ion , affective state and attribute words. Moreover , they can keep a partial representat ion of these words in memory at least as long as a week, suggest ing that the words have actual ly entered the lexicon and have not merely been memor ized a s bits of information unconnec ted to prior knowledge. There s e e m s to be no dif ference in 'fast m a p -ability' of monosy l lab ic and mult isyl labic words. A dif ference w a s found to exist in the amount of information retained b a s e d on the domain of new adject ives. Th is effect may have been c a u s e d by the fact that the chi ldren studied were p red isposed to pay attention to certain attributes of objects (e.g. s h a p e ) more than others (e.g. texture) or by the fact that the chi ldren knew more vocabulary about the s h a p e doma in than other doma ins and thus were able to narrow down the mean ing of the new word in this doma in more efficiently than the others. T h e most vu lnerable aspec t of the fast mapping p rocess , regard less of word type, appears to be the retrieval of phonolog ica l information for spon taneous product ion of the new word. Chi ldren incorporate non- l ingu is t ic information with l inguistic information to form hypotheses about the mean ing of the unknown word and may in fact rely more heavi ly on non- l ingu is t ic information to form these hypo theses . They overwhelmingly assoc ia te the new word with an object/attribute they have no n a m e for themse lves . In He ibeck and Markman 's 1987 study only 2 of 150 subjects behaved as if the new word referred to the familiar object for which they a l ready had a name. In order to be able to fast map, one must have s o m e knowledge of the syntax of the language , even if it is as rudimentary as word order, from which to postulate the grammmat ica l c lass of the word used . T h e tendency to a s s u m e that a new word refers to s o m e new aspec t of an object a s wel l a s to a s s u m e that terms are mutually exc lus ive has been noted in severa l fast mapping s tud ies (D ick inson, 1984; He ibeck and M a r k m a n , 1987). Th is principle is a lso the bas i s of recent theor ies of lexical acquis i t ion s u c h as those p roposed by Barrett ( 1978) and C la rk (1983). Fast 11 mapp ing s e e m s to entail s o m e set of cognit ive and language m e c h a n i s m s which can be appl ied to l inguistic p rob lems. A s the chi ld grows older and matures the component p r o c e s s e s presumably a lso mature, al lowing the chi ld to so lve increasingly complex l inguistic p rob lems. Thus chi ldren are able to fast map in var ious semant ic and syntact ic doma ins (He ibeck and M a r k m a n , 1987; R ice and W o o d s m a l l , 1988), in var ious contexts (R ice and W o o d s m a l l , 1988; D ick inson, 1984) and more than one word at once (R ice and W o o d s m a l l , 1988). A s can be s e e n from Tab le I, in spite of severa l recent s tud ies the picture of fast mapping abil i t ies is c loudy. C l e a r di f ferences in fast mapping ability with age cannot be s e e n , perhaps as a result of sma l l samp le s i z e s in each age group in s o m e studies. However , there do s e e m to be some procedure related di f ferences, espec ia l ly evident in the 3 - y e a r - o l d ' s data . It is difficult though to be. Tab le I. Pe rcen tage correct in var ious fast mapping studies. A g e 2 yrs 3 yrs 4 yrs 5 yrs S tudy C o m p P rod C o m p Prod C o m p P rod C o m p P rod Do l laghan (1985) 7 1 % 57% 83% 0% 9 1 % 4 5 % 7 1 % 7 1 % [objects] (10 subjects) (7 subjects) (11 subjects) (7 subjects) He ibeck and Markman 6 1 % 13% 6 1 % 22% 6 1 % 19% [attributes] (1987) (27 subjects) (30 subjects) (26 subjects) R i ce and W o o d s m a l l (1988) 44% object 70% object [objects and attributes 50% attribute 62% attribute in complex context] (27 subjects) (34 subjects) 12 sure that d i f ferences in per formance ac ross s tud ies are c a u s e d by di f ferences in p rocedures and not d i f ferences in samp le s i ze and se lec t ion. 1.3 F A S T M A P P I N G A N D L A N G U A G E A C Q U I S I T I O N Fas t mapp ing would s e e m to be s o m e set of quick prob lem solv ing m e c h a n i s m s which yield information n e c e s s a r y for the e lementary comprehens ion and product ion of words . S tud ies of chi ldren with a spec i f ic language impairment sugges t that s u c c e s s in performing fast mapping tasks may be related to s u c c e s s in overal l language learning. T h e s e studies have shown that chi ldren with a spec i f ic language impairment perform signif icantly worse on fast mapp ing tasks than do a g e - or M L U - matched normal language peers (Do l laghan, 1987; R i c e , Buhr and Nemeth , 1989; Ape l and K a m h i , 1987). L a n g u a g e impaired chi ldren require signif icantly more exposu res to a word than a normal chi ld before they can p roduce it, and even then they have difficulty retrieving the word after a half hour p a u s e (Apel and Kamh i , 1987). This correlat ion cannot prove that poor fast mapping skil ls c a u s e language de lays . It may be that s o m e underlying deficit c a u s e s both the genera l language impairment and the lack of fast mapp ing ski l ls. Never the less , the studies of language impaired chi ldren support the v iew that the fast mapping paradigm measu res s o m e set of ski l ls, either directly or indirectly, that is important in natural language learning. T h e study of fast mappp ing thus opens up a new w a y of studying language learning capabi l i t ies and perhaps even the m e c h a n i s m s of learning themse lves . 1.31 S e c o n d Language Learn ing A n aspec t of language learning that fast mapping could potential ly enl ighten is that of s e c o n d language acquisi t ion in chi ldren who have a l ready attained a firm grasp of their first language before 13 being e x p o s e d to their s e c o n d . There are severa l d i f ferences between the acquis i t ion of a first l anguage ( L l ) and a subsequen t s e c o n d language (L2). First, L2 learners a l ready have one language sys tem in p lace. A s a result of this they are implicitly famil iar with the notion of sounds represent ing objects and ideas. They a l ready know what a word is and that words can be combined to form sen tences . They a l ready have words for the most c o m m o n objects and ideas in their l ives and may be resistant to learning new words for the ' same thing'. In many c a s e s the L2 learner does not have to s p e a k L2 in order to be understood and can fall back on L l when necessary . This may be espec ia l l y prevalent in the c a s e of E n g l i s h - s p e a k i n g North A m e r i c a n s who are explicit ly taught a s e c o n d language in an unnatural c l ass room context. L2 is not n e c e s s a r y for communicat ion in s u c h c i r cumstances . S e c o n d , L2 learners are not undergoing the rapid and complementary cognit ive and neuro logica l deve lopment that L l learners do in the first three years of life. Op in ions vary concern ing the deve lopment of the auditory sys tem in infants, but there is a possibi l i ty that chi ldren younger than 2 yea rs of age have an auditory sys tem that is still maturing. Thus the quality of the acoust ic information reaching an L l learner early in her/his language exper ience may be quite different from that of an L2 learner with a mature and stable auditory sys tem. Infant L l learners are also deve lop ing s u c h cognit ive notions as object pe rmanence whi le they are learning their language. T h e s e concep ts are a l ready in p lace for the L2 learner. Th i rd , L2 learners, no matter how young , have concep ts about the culture involved with L2 and can s e e the react ions of re lat ives/peers to that culture when they s p e a k the language. If these react ions are negat ive they can lead to a situation where the L2 learner d o e s not want to learn the language s ince s /he will then be perce ived as 'snobb ish and s i ssy ' or ' lower c lass and vulgar'. Is it real ist ic, g iven these di f ferences between first and s e c o n d language acquis i t ion, to expect fast mapp ing to play a role in the acquisi t ion of a second language? I think that it is. 14 T h e first di f ference ment ioned w a s that L2 learners a l ready have a language in p lace . G iven that it is poss ib le to acquire an L2, the fact that the speake r knows L I a l ready doesn ' t tell us much about how L2 is learned. L2 learners have had previous exper ience extracting meaning from the non- l ingu is t ic env i ronment and the conversat ional context. They have a l ready gone through the p rocess of extract ing and segment ing str ings of sounds into meaningful units in one language. P e r h a p s they deve lop more mature lexical learning strategies for s e c o n d language learning, but the phonolog ica l and morphologica l sys tem of the new language still must be learned. The p rocess of learning a s e c o n d language has been the subject of much deba te 3 but it s e e m s intuitively c lear that the ability to quick ly map information about the mean ing , use and form of new words into memory wou ld be a s useful in s e c o n d language learning as in first language acquis i t ion. T h e s e c o n d di f ference be tween young LI learners and slightly older L2 learners ment ioned above w a s the incredible amount and rate of cognit ive and neurological deve lopment occurr ing while chi ldren learn their first language but not whi le learning their s e c o n d . Thus L2 learners have better conceptua l and perceptual resources avai lab le to them than L I learners. If we a s s u m e that fast mapp ing doesn ' t occur in LI learners under 2 years of age b e c a u s e of this lack of resources , then we cou ld postulate that fast mapping may actual ly p lay a bigger role in L2 learning than LI s ince it wou ld begin earl ier in chi ldren's L2 exper ience than in their L I exper ience. Another di f ference between the two types of language learning is the poss ib le effect of the attitudes of relat ives, peers and the child her/himself with respect to the culture assoc ia ted with the s e c o n d language and that assoc ia ted with the first language. If the L2 culture is negatively va lued, or va lued less than LI culture, the child may never ach ieve f luency and/or a lways retain a 'foreign' accen t regard less of her/his true ability to learn. This is espec ia l ly true in the c a s e of minority groups ' submerged ' in a larger language group. T h e s e minority g roups tend to retain a distinctive 3 For a discussion of theories of second language acquisition see Brown, 1987. 15 accen t /g rammar in their L2, not necessar i l y in rejection of the L2 culture, but to keep their own identity as a group of L l speake rs (Wardhaugh, 1986). This di f ference between L l and L2 learning essent ia l ly amounts to a consc ious ( or unconsc ious) dec is ion to use or not use their full range of language learning ski l ls wh ich may include fast mapp ing. However , if the chi ld d o e s dec ide to learn the L2, it wou ld s e e m reasonab le to suppose that fast mapping wou ld play a part in this learning. 1.32 S e c o n d L a n g u a g e Instruction A fourth and frequently ment ioned dif ference between L l and L2 learning is that L2 is often taught explicit ly instead of learned implicitly. But this is not a lways the c a s e . F rench Immersion is a foreign language learning situation in which the Engl ish speak ing majority culture students are taught the schoo l curr iculum in F rench . F rench Immersion is a C a n a d i a n invent ion which started a lmost s imul taneous ly in Toronto and Montreal in 1965. S i n c e the publ icat ion of numerous studies of the pilot c l a s s e s showing no negat ive long term a c a d e m i c effects on the students as a result of teach ing in a s e c o n d language (Lambert and Tucker , 1972; S w a i n , 1972), F rench Immersion has rapidly b e c o m e a high prest ige form of educat ion in C a n a d a assoc ia ted with high income, w e l l -educa ted famil ies. In this sys tem of instruction the teacher is fluent in the first language and culture of the chi ldren and these things are never denigrated. Indeed the wor th iness of L l is demonst ra ted by the fact that, by G r a d e 6, 50% of the curr iculum is taught in Eng l i sh , not F rench . In this French speak ing situation any communicat ion in F rench is app lauded and encou raged , even if not 'correct'. T h e teacher modi f ies her/his output to the students, making it comprehens ib le by speak ing slowly with heavy intonation, lots of repetition and paral inguist ic cues (Cummins , 1984) 4. In the tradit ional s e c o n d language c lass room, with its lists of vocabu lary to be memor ized , 4 This is completely unlike the 'submersion' experience of a child whose L l is unknown to the teacher, blamed for his/her academic problems and not recognized as valuable. 16 pronunciat ion drills and explicit explanat ions of grammat ica l rules, it is conce ivab le that fast mapping would have a negl igible role in L2 learning. In such a situation exper iment ing with the language is not a priority; everything is taught, not 'd iscovered ' . In the French Immersion sett ing, however, chi ldren are encou raged to exper iment with F rench , to g u e s s what things m e a n and how best to exp ress themse lves . In such a communicat ive learning envi ronment the ability to perform fast mapp ing tasks wou ld probably help the L2 learners cons iderably . 1.33 S u m m a r y In any s e c o n d language learning situation (i.e. language learning after the first language is functional) the L2 must be teamedsomehow. The chi ldren are cognit ively and neurological ly more mature than w h e n they learned their first language, but what must be learned in order to know a language has not changed . If this L2 learning were to take p lace in a natural ist ic sett ing such as F rench Immersion, it s e e m s realist ic to expect s o m e sort of fast mapp ing to take p lace as wel l . Chi ldren impaired in their first language are signif icantly poorer at producing (Dol laghan, 1987) and understanding (R ice , Buhr and Nemeth , 1989) new words presented only a few t imes than are their a g e - and M L U - m a t c h e d peers . G i ven this correlat ion between first language acquisi t ion s u c c e s s and fast mapp ing , and the fact that fast mapping could be involved in the learning of any language , it s e e m s logical to investigate a connect ion be tween seconc f language acquisi t ion s u c c e s s and fast mapp ing . 1.4 S U C C E S S IN F R E N C H I M M E R S I O N R e g a r d l e s s of the naturalist ic exposure to French and high prest ige of F rench Immersion s o m e chi ldren do not do wel l in this program. This c a u s e s a great dea l of e x p e n s e and confusion to all conce rned s ince a large amount of time is invested in educat ing a chi ld in F rench Immersion, time 17 spent by the chi ld, teachers and parents. If it is eventual ly d e e m e d best for s o m e chi ldren to d iscont inue the F rench program and enter the Eng l ish program before G r a d e 3 (which is when Eng l i sh l i teracy training beg ins in F rench Immersion), then these chi ldren may perce ive themse lves a s fai lures in F rench and will f ind themse lves as much as two years beh ind in Eng l i sh , s imply b e c a u s e they have never been taught Engl ish l i teracy ski l ls. In an effort to avo id such diff iculties, many investigators have tried to f ind predictors for s u c c e s s in F rench Immersion that could be u s e d to sc reen out chi ldren at risk for p rob lems from the very beginn ing of the program. S u c h efforts have met with limited s u c c e s s , pe rhaps b e c a u s e of s o m e confus ion as to what consti tutes ' s u c c e s s ' in F rench Immersion and a lso confus ion about the .goal of this predict ion. Intuitively, success fu l s tudents in F rench Immersion wou ld s e e m to be a c l ass of chi ldren who do as wel l academica l l y in F rench a s they wou ld have in Eng l i sh , and who a lso have b e c o m e fluent in F rench . 'Fa i lure ' wou ld then occur if a student fai led to b e c o m e fluent in F rench or per formed worse in c l a s s e s taught in F rench than in those taught in Eng l i sh . But this is not the definition of s u c c e s s used by researchers , or by. the schoo l s or the parents involved. F rench Immersion is expec ted to create students who are literate in F rench and who unders tand spoken F rench , but not necessar i l y peop le w h o c a n speak F rench . Indeed, a report on one of the earl iest s y m p o s i u m s on bi l ingual educat ion in C a n a d a stated that F rench Immersion w a s not in tended to create f luency, but that after approximate ly three months in a French environment, g raduates of F rench Immersion p rograms shou ld be " reasonab ly fluent" (Swa in , 1972:77). F rench Immersion has a lso become somewha t elitist, leading to c l a s s e s where the average IQ and ach ievement level is s o high that a chi ld performing at an age-appropr ia te a c a d e m i c level may be at the bottom of the c lass and perce ived as a 'fai lure' (Cummins , 1984). Further confus ion has resul ted from failure to dist inguish be tween criteria which can be used 18 to predict genera l a c a d e m i c s u c c e s s ( including that involved in F rench Immersion) and criteria which can be used to predict wh ich chi ldren would actual ly be hand icapped academica l l y by instruction in F rench as o p p o s e d to Eng l i sh . 1.41 Pred ic tors of S u c c e s s in F rench Immersion It is apparent then that ' s u c c e s s in F rench Immersion' is not necessar i l y the s a m e as ' s u c c e s s in learning F rench as a s e c o n d language ' and that a predictor of s u c c e s s as portrayed in the literature d o e s not necessar i l y differentiate between chi ldren who wou ld do poorly in any program and those who wou ld only do poorly in F rench . T h e results of s tud ies search ing for predictors of ' s u c c e s s ' reflect these dispar i t ies. Tr i tes (1976) deve loped a Tactua l Per fo rmance Test requiring chi ldren to p lace var ious s i zes and s h a p e s of b locks into a form board with either hand whi le bl indfolded. He found that chi ldren having difficulty in F rench Immersion, or who had actual ly d ropped out into the Engl ish program, were signif icantly poorer at the T P T than those chi ldren still in F rench Immersion. Per fo rmance difficulty on the T P T w a s postulated to be the result of a maturat ional lag in the temporal lobe regions of the bra in, regions which are wel l known centres of language learning. Trites c la imed that chi ldren with difficulty in F rench Immersion are a group distinct from any other chi ld deve lopmenta l d isorder and that these chi ldren would do-better in their native language than in F rench (Trites and Pr i ce , 1979). Tri tes did not provide any ev idence from C T - s c a n s for his c la im, nor speci fy which hemisphere this lag occurs in. A n examinat ion of Tr i tes' da ta by C u m m i n s (1984) s h o w e d that the chi ldren who d ropped out of F rench Immersion into Eng l ish did not improve their academic ach ievement to expec ted levels for their age when taught in their own language, suggest ing that whatever deficit the T P T w a s picking up w a s not speci f ical ly c a u s e d or exacerba ted by instruction in F rench . T h u s , whi le the T P T may have predicted failure in F rench Immersion, the failure does not 19 s e e m to be spec i f i c to L2 instruction. If anything, the T P T s e e m s to identify chi ldren w h o are poor c l ass room ach ievers regard less of language of instruction. R e s e a r c h e r s have a lso studied the personal i ty traits of chi ldren in F rench Immersion. Swa in and Burnaby (1976) asked teachers to subject ively rate their s tudents on a variety of personali ty character is t ics (e.g. cl inging to one 's own opin ion, independence) . They found that "qu ickness at g rasp ing new concep ts " correlated posit ively with severa l object ive m e a s u r e s of F rench ability, whi le "per fect ion ism" corre lated negatively with pronunciat ion (Swain and Burnaby , 1976). Th is type of predict ion is of l imited use fu lness as little is known about how personal i ty deve lops over t ime. A chi ld who is " independent" in Kindergar ten may b e c o m e "anx ious" by G r a d e 3. A l s o , the teachers ' subject ive ratings may be unintentionally b iased by whether or not they like the chi ld. IQ has been found to be a good predictor of a c a d e m i c per formance in F rench Immersion and of acquis i t ion of l i teracy ski l ls in F rench ( G e n e s e e , 1976). Th is is not surpr is ing s ince the prediction of a c a d e m i c s u c c e s s in genera l is essent ia l ly the purpose of IQ sco res . But IQ d o e s not predict conversa t iona l , oral F rench ski l ls (Cummins , 1984) s o it cannot be sa id to be a predictor of s u c c e s s in F rench Immersion as a who le , merely one aspec t of it. Morover , in a crit ical review of "exist ing research related to the suitability of F rench Immersion for d i sadvan taged and minority group ch i ldren" G e n e s e e (1976) sugges ts that a chi ld 's motivation to learn the language and attitude toward the culture invo lved with the language interacts with her/his IQ to g ive a good predictor of ' success ' in F rench Immersion. T h u s chi ldren with high motivation and low IQ's might wel l be sui table for F rench Immersion programs. Another predictor of s u c c e s s in F rench Immersion that has been invest igated is a previously d i agnosed language disorder. Bruck (1978,1982), G e n e s e e (1976) and C u m m i n s (1984) are of the op in ion, b a s e d on sol id we l l - rep l ica ted research , that language impai red chi ldren do no worse in F rench Immersion than they do in mainst ream Engl ish c l a s s e s . Be ing taught in a s e c o n d language 20 d o e s not s e e m to exacerba te their prob lem or retard their a c a d e m i c ach ievement and actual ly may raise their self es teem (Bruck, 1978). F rom one perspect ive then, it cou ld be a rgued that language impaired chi ldren wou ld benefi t f rom inclusion in F rench Immersion programs s ince they could gain s o m e compe tence in a s e c o n d language in these programs. Opponen ts of this v iew might argue that, whi le F rench Immersion doesn ' t d iminish the a c a d e m i c per formance of language impaired ch i ldren, the fact remains that language impaired chi ldren do poorly in schoo l , no matter what language they are taught in. F rom this perspect ive language impai red chi ldren s e e m unable to s u c c e e d in F rench Immersion programs, as ' s u c c e s s ' is now conce i ved . Indeed it has been found that there is a signif icant posit ive correlat ion be tween Engl ish and F rench l i teracy ski l ls in all chi ldren in F rench Immersion, suggest ing that there is s o m e underlying 'genera l ability' - or disabil i ty - in Language that will mani fest itself regard less of the speci f ic l anguage the speake r is us ing (Cummins , 1984). Language impaired chi ldren wou ld represent the far end of this ability cont inuum. Mov ing inwards towards the m e a n , there may a lso be a group of chi ldren w h o s e genera l language ability is adequate for a c a d e m i c ache ivement in L l but insufficient to handle c l ass room instruction in L2. Whether s u c h a group of chi ldren exists is unknown s ince there has been surpr is ingly little research done to invest igate this quest ion in the F rench Immersion sett ing. Mos t of the exist ing literature in F rench Immersion a s k s whether the program has affected the chi ldren 's a c a d e m i c ski l ls and/or personal i ty, not v ice ve rsa . T h e above review represents virtually all the studies that have tried to d iscover what character is t ics the chi ld br ings to the French Immersion exper ience that may affect her/his per formance therein? 1.42 S u m m a r y F rench Immersion is a s e c o n d language learning situation in a schoo l context but the language teaching is implicit, much like the situation for natural language learning. ' S u c c e s s in 21 F rench Immersion ' has c o m e to incorporate a c a d e m i c s u c c e s s a s wel l a s s u c c e s s in learning F rench . Th is has made the sea rch for predictors of s u c c e s s in F rench Immersion confus ing, s ince a c a d e m i c s u c c e s s is not a prerequisi te to learning a s e c o n d language ( C u m m i n s , 1984; G e n e s e e , 1976); but not understanding the subject matter, b e c a u s e it is p resen ted in F rench (the s e c o n d language) , can limit a c a d e m i c s u c c e s s . IQ h a s proven to be a good predictor of the literate and a c a d e m i c portion of F rench Immersion (Cummins , 1984). A previously d iagnosed language d isorder in the first language s e e m s to a lso impair the learning of a s e c o n d language (Bruck, 1978; G e n e s e e , 1976; C u m m i n s , 1984). Interacting with both of these a reas of genera l cognit ive ability and language prof ic iency are personal i ty and cultural factors which vary widely from child to chi ld. 1.5 T H I S S T U D Y A s a result of the high price pa id in t ime, a c a d e m i c progress and att i tudes toward schoo l by s tudents , parents and educators in educat ing a chi ld in F rench Immersion only to find that s /he would be better off in an Eng l ish c l ass room, a predictor is needed to dec ide early in a chi ld 's life whether or not s /he is a good candidate for F rench Immersion. A 'good cand idate ' wou ld be s o m e o n e who wou ld be able to learn F rench and , as a corol lary, not be hand i capped academica l l y by being instructed in F rench . The present study is an examinat ion of the L2 learning aspec t of French Immersion. In part icular this paper will explore the possibi l i ty that fast mapping may be a predictor of s u c c e s s in acquir ing a s e c o n d language. T h e fast mapp ing ski l ls of a group of chi ldren in a Pr imary F rench Immerison Program were tested and then ana lysed to s e e if there w a s any relat ionship be tween fast mapp ing ski l ls and 1) F rench reading comprehens ion , 2) F rench oral comprehens ion , 3) F rench oral exp ress ion , 4) teacher ratings of overal l a c a d e m i c ability and 5) whether or not the chi ld w a s still enrol led in the Pr imary 22 French Immersion program. It was expected that the "underlying general [language] ability" mentioned by Bruck (1978) would be captured by the three measures of French ability and that these in turn would correlate with fast mapping skill. If a child has specific problems with the literacy aspect of language, then the two oral measures would be expected to correlate with fast mapping while the reading measure would not. If those children who are now in English programs after 'failing' in French Immersion have significantly lower fast mapping scores than those children still in the program, this would imply that fast mapping is a predictor of French Immersion success and taps into whatever faculties school administrators base such decisions on. 23 C H A P T E R 2 M E T H O D 2.1 O V E R V I E W Students present ly and formerly enrol led in an Ear ly F rench Immersion program at an e lementary schoo l in Brit ish C o l u m b i a were tested on their fast mapp ing ski l ls using a procedure modi f ied from Dol laghan (1985). T h e chi ldren were e x p o s e d to an unfamil iar object, for which they d id not yet have a name from the electr ical supp l ies department of a hardware store. In the course of a hiding g a m e the nonsense name of this object w a s ment ioned. After the French  C o m p r e h e n s i o n Tes t w a s administered as a distractor the chi ldren were tested to s e e whether they knew what the word referred to and if they could produce the word. T h e results of the fast mapping test were compa red to the chi ldren's results on : i) the Tourond tests d iaqnost iques de lecture, a s tandard ized F rench reading test which is used in part to measure their s u c c e s s in F rench , ii) teacher ratings of overal l a c a d e m i c ability, iii) teacher ratings of oral F rench ability and iv) TheFrench  C o m p r e h e n s i o n Tes t , an oral F rench comprehens ion test. 2.2 S U B J E C T S Sub jec ts were G r a d e 2 and 3 students with a wide range of F rench ski l ls. E ighteen students were in G r a d e 2 F rench Immersion and had been e x p o s e d to French s ince Kindergarten (3 years). S e v e n t e e n students were in G r a d e 3 French Immersion and had been e x p o s e d to 4 years of F rench. F ive s tudents were former French Immersion students who had swi tched to a mains t ream Engl ish p rogram and were in G r a d e 3. T h e s e students had each been in F rench Immersion for 3 years . All subjects spoke Eng l ish a s a first language. S i n c e this w a s not intended to be a study of the deve lopment of fast mapp ing ski l ls, but rather a sea rch for the relat ionship between fast mapping and the deve lopment of a s e c o n d language, no 24 attempt w a s made to have equa l numbers of students of var ious ages . T h e subjects inc luded e leven 7 - y e a r - o l d s , twenty 8 - y e a r - o l d s and nine 9 - y e a r - o l d s . 2.3 P R O C E D U R E Al l chi ldren involved in the study were tested individually in a quiet room al located by the schoo l . S e s s i o n s cons is ted of a per iod of conversat ion to acquaint the chi ld with the investigator and the fol lowing five tasks . Al l instructions were g iven in Eng l i sh . T h e entire procedure took approx imate ly 15 minutes. E a c h chi ld per formed the tasks in the s a m e order and w a s tape recorded. 2.31 Exposu re T a s k O n a table be tween the chi ld and the examiner were p laced three objects ( p e n , fork and X) and three p laces to put the objects ( a box, a bowl and s o m e t issue paper) . The chi ld w a s instructed first to hide the pen , then the fork and t hen—when there w a s no cho ice of object or hiding p lace left— - t h e unknown object. If, after 5 s e c o n d s , the chi ld did not hide the object, the examiner sa id " Y o u hide it". T h e name of the object w a s ment ioned only once . It w a s necessa ry that the chi ld 's only exposu res to the nonsense words shou ld be during the s e s s i o n . To avo id the possibi l i ty of chi ldren d i scuss ing the procedure with each other and inadvertently telling e a c h other the word they would have to remember a variety of target C V C nonsense words were u s e d (see Tab le 11). Chi ldren were tested individually, c l ass by c lass . E a c h chi ld w a s ass igned a number 1 through 6 and the target word from Tab le II cor respond ing to that number. Ch i ld ren were then tested in order of numbers (i.e. first the chi ld with word #1, then word #2, etc.) 25 Tab le II. List of target words and their distractors. T A R G E T D I S T R A C T O R 1. 2. 3. 4. 5. 6. [kut] [git] [put] [kit] [bud] [kid] [zut], [pid] [zit], [bud] [zut], [kid] [zit], [pud] [shid], [put] [shud], [pit] E a c h word conta ined a high, tense vowel ([i] or [u]). Ve la r and labial s tops occur word initially more often than in any other posit ion in the s p e e c h of chi ldren in G r a d e s 2 and 3 (Mader , 1954 as ci ted in Shr iberg and Kent, 1982). To ensure that the words were not so unusua l that the chi ldren wou ld pay extra attention to them, or so hard that s o m e chi ldren would be unable to pronounce them, all of the n o n s e n s e labels conta ined a velar or labial stop a s the first consonant . A lveo lar stops occur word finally more often than in any other posit ion in this age range (ibid) and so the final consonan ts were all a lveolars. 2.32 F rench C o m p r e h e n s i o n Test T h e F rench C o m p r e h e n s i o n Tes t is an oral F rench comprehens ion test deve loped by the Ontar io Institute for S tud ies in Educat ion speci f ical ly for chi ldren in F rench Immersion programs. Ch i ld ren were tested to determine, independent of poss ib le teacher b ias , what their F rench oral comprehens ion w a s with respect to other chi ldren in F rench Immersion situations. Th is a lso served a s a distractor task be tween the exposure and testing tasks. The chi ldren were all admin is tered the 'Words and S e n t e n c e s ' subtest of the F rench C o m p r e h e n s i o n Test . They were a s k e d to point to the picture that matched best the F rench word or sen tence spoken by the examiner . 26 2.33 C o m p r e h e n s i o n T a s k Two new unfamil iar objects were added to the three objects u s e d in 2.31 above . From the array of two known and three unknown objects, the chi ld w a s a s k e d to give the examiner the pen , the X , and then the fork. 2.34 Product ion T a s k T h e chi ld w a s then a s k e d to name each object as the examiner held it up. Encouragement w a s g iven to g u e s s after 5 s e c o n d s with no attempt. 2.35 Recogn i t ion T a s k Th is task w a s g iven to those chi ldren who fai led to attempt the label l ing task. T h e chi ld was a s k e d "Is this an X , a Y or a Z ? " Al l words were n o n s e n s e words maximal ly distinct from the target nonsense C V C (see Tab le II). 2.36 Li teracy Tes ts Al l chi ldren involved in the study had taken the Tourond tests d iagnost iques de lecture at the end of the prev ious a c a d e m i c year. T h e s e are a set of F rench reading tests des igned for Early F rench Immersion students by the Ontar io Institue for S tud ies in Educa t ion . The principal of the schoo l and the F rench Learn ing Ass is tan t teacher made these sco res avai lable to the researcher . T h e subtest s c o r e s u s e d in the present study were from Par ts I and II of both the G r a d e 1 and Grade 2 tests (performed by the chi ldren currently in G r a d e s 2 and 3, respect ively) . Part I (Word Recogni t ion) tested the chi ldren's ability to dec ipher the g rapheme /phoneme relat ionship in French by ask ing the chi ldren to circle the written French word that the teacher sa id . Part II (Word Meaning) tested the student 's comprehens ion of written French by ask ing them to circle the w o r d , from a field 27 of four, that matched the picture. 2.37 Teacher Ratings In addition to these direct measures of performance, the classroom teachers of all the children involved in the study were asked to rate each child's math skills and oral French skills with respect to their classmates using a four point rating system (see Appendix A). Areas of classroom instruction in the age range used in this study are commonly divided between 'language' and 'maths/sciences'. It was hoped that a measure of the child's math skills would indicate general, non-language academic ability. Teachers were asked to rate each child's oral French ability, since no standardized tests of oral French production were available. 28 C H A P T E R 3 R E S U L T S 3.1 F A S T M A P P I N G T h e purpose of this study w a s to ascer ta in what, if any, relat ionship there is between fast mapp ing and s u c c e s s in F rench Immersion (FI) p rograms. In order to answer this quest ion chi ldren either currently or formerly enrol led in Ear ly F rench Immersion were g iven a set of fast mapping tasks and five m e a s u r e s of s u c c e s s in French Immersion. T h e s e measu res were a teacher rating of oral F rench product ion, a F rench reading test, a teacher rating of math prof ic iency and an oral F rench comprehens ion test. The fifth measure of s u c c e s s w a s whether or not the student w a s still enro l led in F rench Immersion. T h e subjects involved inc luded G r a d e 3 chi ldren no longer enro l led in FI, G r a d e 3 chi ldren in FI and G r a d e 2 chi ldren in FI. Al l of the chi ldren ach ieved 100% s u c c e s s in both the exposure and comprehens ion tasks . Da ta from these tasks will therefore not be ana l yzed further. In the production task 27 subjects p roduced all three p h o n e m e s in the target word , 7 p roduced only two p h o n e m e s and 6 subjects were unable to produce any. 3.11 Fas t Mapp ing and G r o u p Membersh ip T h e relat ionship be tween these results and cont inued enrol lment in F rench Immersion was tested us ing a c h i - s q u a r e ana lys is . The distribution of fast mapp ing product ion sco res per grade can be found in Tab le III. No signif icant relat ionship between group membersh ip and fast mapping product ion w a s found {%= .79, df=4,O(=.05). A s can be s e e n in Tab le III, three of the chi ldren no longer in F rench Immersion were ab le to provide all three p h o n e m e s of the target word while three of the chi ldren still in G r a d e 3 F rench Immersion complete ly fai led the task. There a lso w a s no 29 signif icant di f ference in fast mapping ability with increasing number of yea rs of exposure to a second language. Tab le III. Distr ibution of fast mapping product ion sco res by instructional group. G r o u p G r a d e 2 G r a d e 3 G r a d e 3 F rench Immersion French Immersion Engl ish Fas t Mapp ing (n=18) (n=17) (n=5) S c o r e Fai l (0 or 1 2 3 1 of 3 phonemes ) Part ia l P a s s 3 4 l (2 of 3) P a s s 13 10 3 (3 of 3 phonemes ) 3.12 Fas t Mapp ing and Ora l F rench Product ion A s s tandard ized measu res of spoken French were unavai lab le , teachers were a s k e d to rate their s tudents in this a rea (see Append ix A). T h e s e four point ratings were used to investigate the relat ionship be tween fast mapping product ion per formance and oral F rench product ion. C r o s s -tabulated distr ibutional da ta for these two measu res are presented in Tab le IV. C h i - s q u a r e analys is revea led no signif icant relat ionship be tween these two measu res (75=2.1, df=6#=.05). Ch i ld ren rated 30 poorly in F rench product ion ski l ls per formed no better or worse in fast mapping than those rated wel l . Tab le IV. Distribution of fast mapping product ion sco res by oral F rench product ion ski l l . Rat ing S c a l e 1 (worst) 2 3 4 (best) Fas t Mapp ing S c o r e Fai l 1 2 1 1 Part ia l P a s s 1 2 2 2 P a s s 1 7 7 8 3.13 Fas t Mapp ing and G e n e r a l A c a d e m i c Abil ity A s a measu re of each chi ld 's overal l , n o n - l a n g u a g e a c a d e m i c ability teachers rated the chi ldren's math abil it ies us ing the s a m e four point sys tem as above . A c h i - s q u a r e ana lys is showed no signif icant relat ionship be tween fast mapping and math ability o£=9.l, df=6C<=.05). A s can be s e e n in Tab le V , none of the chi ldren rated a s poorest in math ski l ls fai led the fast mapping task. 31 Tab le V. Distribution of fast mapping product ion sco res by teacher rating of math abil i t ies. Rat ing S c a l e 1 (worst) 2 3 4 (best) Fas t Mapp ing S c o r e Fai l 0 3 2 1 Part ia l P a s s 1 0 5 2 P a s s 1 3 12 10 T h e s e results show no consistent change in fast mapp ing skil l related to cont inued inclusion in F rench Immersion, length of exposure to a s e c o n d language, product ive language ski l ls in the s e c o n d language or overal l a c a d e m i c ability. The final two measu res of F rench Immersion s u c c e s s , the Tourond reading test and the F rench C o m p r e h e n s i o n Test , were s tandard ized tests yielding an ordered cont inuum of percen tages and raw sco res . It w a s poss ib le therefore to ana lyse these data with a more powerful parametr ic procedure. 3.14 Fas t Mapp ing and W o r d Recogni t ion Par t I of the Tourond tests d iaqnost iques de lecture (Word Recogni t ion) tests the students ' g rasp of the g rapheme/phoneme relat ionship by having the chi ldren circle the written word or syl lable that the teacher s a y s . T h e S p e a r m a n ' s rank correlat ion coeff icient for the relat ionship between the fast mapp ing product ion task and this subtest w a s found to be quite low (r= .31) s, demonstrat ing a 5This Spearman coefficient has not been corrected for ties, the corrected value would be even lower. 32 w e a k correlat ion be tween per formance on the two tests. Ch i ld ren who were good at decod ing written F rench were not necessar i ly good at fast mappp ing . 3.15 Fas t Mapp ing and W o r d Mean ing Part II of the Tourond (Word Meaning) tested the chi ldren's understanding of the meaning of written F rench words be ask ing them to choose the correct word (one of four) to descr ibe a picture. T h e S p e a r m a n ' s rank correlat ion coeff icient for the relat ionship be tween this reading comprehens ion test and the fast mapping product ion task a lso indicated a very w e a k correlation be tween s c o r e s (r= .26) 6. 3.16 Fas t Mapp ing and Ora l F rench C o m p r e h e n s i o n The F rench C o m p r e h e n s i o n Tes t invest igated the subject 's oral F rench comprehens ion . Ch i ld ren were a s k e d (in Engl ish) to point to the picture (one of four) that best matched the French word or sen tence spoken by the examiner . Ana lys i s of these da ta also s h o w e d a very smal l correlat ion to the fast mapping sco res (r= .28) 7. 3.17 Fas t Mapp ing and A g e S i n c e none of the five original measu res were found to be even moderate ly assoc ia ted with fast mapp ing abil i t ies, the possibi l i ty that fast mapping may improve with age w a s invest igated using a c h i - s q u a r e ana lys is . No signif icant relat ionship between age and fast mapping w a s found ic=7.6, df=4CX=.05). A s can be s e e n in Tab le VI, the 7 - y e a r - o l d s per formed no better or worse on the fast mapp ing product ion task than did the 9 - y e a r - o l d s . 6 see footnote #4. 7This Spearman coefficient has not been corrected for ties, the corrected value would be even lower. 33 Tab le VI. Distribution of fast mapping product ion sco res by age. A g e 7 years 8 years 9 years (11 subjects) (20 subjects) (9 subjects) Fas t Mapp ing S c o r e Fai l 0 5 0 Part ia l P a s s 4 4 1 P a s s 7 11 8 3.2 R E L A T I O N S H I P S B E T W E E N M E A S U R E S O F F R E N C H P R O F I C I E N C Y Con t inued enrol lment in F rench Immersion, number of years exper ience in a s e c o n d language , judgements of expert ise in math and ability in reading, understanding and/or speak ing F rench were all found to be unrelated to fast mapping product ion ski l ls in this study. T h e s e results ra ise one further quest ion. Wha t is the relat ionship between the var ious measu res of language s u c c e s s ? Shou ld the above results be v iewed as three vers ions of the s a m e relat ionship (or lack thereof) or do they reveal that fast mapping is unrelated to three, separa te aspec ts of F rench prof ic iency, (i.e., reading comprehens ion , oral comprehens ion and oral product ion)? To explore this quest ion S p e a r m a n ' s rank order correlat ion coeff ic ients were calculated for all poss ib le combinat ions of the French prof ic iency measu res (see Tab le VII). 34 Tab le VII. S p e a r m a n ' s rank correlat ion coeff icients for relat ionships be tween var ious measu res of F rench prof ic iency (corrected for ties). O R A L L I T E R A C Y Product ion C o m p r e h e n s i o n W o r d Recogni t ion O R A L C o m p r e h e n s i o n .32 L I T E R A C Y W o r d Recogn i t ion .33 .35 W o r d Mean ing .41* .53* .55* * these values are significantCX=.01. Modera te correlat ions were found between oral comprehens ion and reading comprehens ion (Word Mean ing) (r=.53, corrected for t ies) and between the two reading measu res (r=.55, corrected for t ies) but all other re lat ionships were quite weak . Th is is consis tent with the hypothes is that these m e a s u r e s tapped into different, relatively independent , aspec ts of F rench prof ic iency. Thus , it s e e m s that fast mapp ing , as tested in this study, w a s not assoc ia ted with any of severa l different aspec ts of L2 ability. T h e s e da ta provide ev idence that F rench prof ic iency involves different factors related to i) modal i ty, i.e., oral language vs. print, and ii) l anguage p rocess , i.e., comprehens ion vs. product ion. 35 C H A P T E R 4 D I S C U S S I O N 4.1 F A S T M A P P I N G Cont rary to expectat ions, per formance on the fast mapp ing task d id not show signif icant re lat ionships with any of the language and a c a d e m i c measu res u s e d in this study. Th is may not however , indicate a true lack of relat ionship be tween fast mapp ing and oral or literate s e c o n d language ski l ls but may rather be due to a cei l ing effect. On ly six of the 40 chi ldren tested were unable to p roduce any of the p h o n e m e s in the target word and only s e v e n could p roduce two of the three p h o n e m e s . Al l other subjects per formed all the tasks with no difficulty. O n e poss ib le explanat ion for the cei l ing effect is that the level of fast mapping task used in this study may be fully within the capabi l i t ies of all 7 year olds ( the younges t chi ldren in the study). The re is insufficient da ta at this t ime to make strong c la ims concern ing the deve lopment of fast mapp ing ski l ls with age . A s rev iewed in Chapte r l it is difficult to compare results be tween studies b e c a u s e the research d o e s not i) all centre on the s a m e age range, ii) include comparab le tasks even if they are studying the s a m e age range and iii) include enough subjects within an age range to provide rel iable results (specif ical ly Do l laghan 's study). It is interesting to note however that there is a cons iderab le inc rease in s c o r e s from Dol laghan 's 5 - y e a r - o l d s (71%,71%) to this study's 7 year o lds (100%, 98%) using the s a m e procedure. Th is inc rease may be the result of the deve lopment chi ldren undergo be tween a g e s 5-7 years . A s e c o n d poss ib le explanat ion may be that exposure to a s e c o n d language actually i nc reases one ' s fast mapp ing ability. T h e very act of learning a s e c o n d language may somehow inc rease one ' s fast mapp ing ability, perhaps by forcing a c o n s c i o u s n e s s of the arbitrar iness of the word- re ferent relat ionship. If this is the c a s e , maybe chi ldren who have had s o m e exposure to a s e c o n d language are more actively involved in postulat ing mean ings for unknown words than 36 monol ingua ls of the s a m e age s ince they have been exposed to more new words . Di f ferences in fast mapp ing abil i t ies be tween monol inguals and bi l inguals is beyond the s c o p e of this study but wou ld be an intriguing topic for further research . R e g a r d l e s s of whether the cei l ing effect w a s the result of exposure to a s e c o n d language or a group of sub jects w h o were too o ld for the task, it s e e m s likely from a review of the literature that a h igher level fast mapp ing task wou ld have el icited a wider range of per formance. Compara t ive ana lys is of the tasks u s e d by other researchers offers s o m e indication of the w a y s in wh ich the fast mapp ing parad igm cou ld be altered so as to capture deve lopment dur ing the schoo l years . Two features that could be var ied are the number and context of the target words. In the present study only one target word w a s used and it w a s only spoken twice. Th is word w a s presented in only one f ixed l inguistic and non- l ingu is t ic context. In R ice and W o o d s m a n ' s (1988) procedure chi ldren (27 ,3 -yea r -o lds ; 34, 5 - y e a r - o l d s ) were e x p o s e d to 20 new words . E a c h word was p resen ted five t imes in a variety of contexts in a T V program. R ice and W o o d s m a l l found that object, act ion and attribute words were equal ly amenab le to fast mapping and that 5 - y e a r - o l d s were signif icant ly better at doing this than 3 - y e a r - o l d s . Pe rhaps 7 to 9 - y e a r - o l d s wou ld do even better at this task. R ice and Woodsma l l ' s comprehens ion sco res for the 5 - y e a r - o l d s exposed to object wo rds (70%) are the s a m e a s the results in Do l laghan 's study (71%; s e e Tab le I) but their sco res for the 3 - y e a r - o l d s are cons iderab ly lower than Dol laghan 's . Th is sugges ts that context and number of words may effect comprehens ion per formance in fast mapp ing . S i n c e comprehens ion sco res are a lways higher than product ion sco res in fast mapping s tud ies, it s e e m s likely that there would be an even more marked deve lopmenta l di f ference in production s co res w h e n context and number of words are var ied . Another factor that could be var ied is the semantic field of the target word. The present study used only an object label a s the target. R i ce and W o o d s m a l l (1988) u s e d act ion, object, attribute and 37 affective state words in their p rocedure. They found that affective state words were significantly more difficult to fast map than the other three word types. P e r h a p s the use of this type of word would have e l iminated the cei l ing effect in the present study. Increasing the time lag between exposure and testing cou ld a lso affect per formance on a fast mapp ing task. T h e procedure used in this study tested comprehens ion and product ion of the new word relatively s o o n after the exposure task (10 minutes). D ick inson (1984) found that when 4 to 9 -y e a r - o l d chi ldren were a s k e d to dec ide if a new word w a s a word , and if a new word w a s used correct ly in a sen tence there w a s no signif icant di f ference in their ability to do this between testing done 5-10 minutes after exposure and that done three to s e v e n d a y s later. A p e l , Kamhi and Do l laghan (cited in Ape l and Kamh i , 1987) report, however , that normal language preschoolers demonst ra ted a signif icant dec rease in ability to produce a new word after only a half hour delay be tween exposure and test ing. T h u s a delay of severa l days be tween the exposure and testing tasks might a lso el iminate the cei l ing effect found in this study. Ye t another poss ib le compl icat ing factor in the fast mapping parad igm might be the phonological complexity of the target word. A p e l and Kamh i (1987) found no signif icant di f ferences in fast mapp ing per formance between either language impaired or normal p reschoo le rs e x p o s e d to monosy l lab les and those e x p o s e d to mult isyl lables. B a s e d on this ev idence the use of monosy l lab ic words in the present study may not have been the c a u s e of the cei l ing effect. However , in conjunct ion with s o m e or all of the previously ment ioned factors, phono log ica l complexi ty could inc rease the difficulty of the fast mapping task. 4.11 S u m m a r y T h e purpose of this paper w a s to sea rch for a predictor of s u c c e s s in F rench Immersion programs. Fas t mapp ing , as tested for in this study, w a s not found to correlate with any of the 38 m e a s u r e s of F rench language ability, either oral or written and thus cannot be u s e d to predict French Immersion s u c c e s s . Th is w a s probably due to the fact that most of the subjects could do the tasks, regard less of L2 prof ic iency. If a more complex fast mapping task w a s u s e d s o m e signif icant correlat ion might be found. It remains poss ib le , however , that fast mapping abil it ies are s imply not corre lated to s u c c e s s in F rench Immersion. 4.2 T H E P R O B L E M O F P R E D I C T I O N A n examinat ion of the relat ionship be tween the language m e a s u r e s of F rench prof iciency u s e d in this study revea led that the literacy measu res were moderate ly assoc ia ted with each other, a s were the comprehens ion measu res . But, in genera l , the oral and l i teracy m e a s u r e s were only weak ly assoc ia ted . Th is sugges ts that these tests are sensi t ive to different aspec ts of L2 prof iciency. It is conce ivab le that a chi ld could be fluent in L2 but illiterate in that language, just as this is poss ib le in L I . S tandard i zed reading tests such as the Tourond currently play a large part in educators ' dec i s ions a s to the proper c l ass room p lacement of a chi ld in F rench Immersion. The lack of correlat ion be tween the oral and literacy measu res u s e d in this study s h o w s once again that L2 prof ic iency, like L I , is made up of severa l abil i t ies. S u c c e s s in one modal i ty.e.g. oral language, does not guarantee s u c c e s s in another, e.g. print ( C u m m i n s , 1984). Nei ther is there a s imple relationship be tween language p r o c e s s e s , i.e. comprehens ion vs product ion, within a modali ty. In order to obtain a true picture of the language abil it ies of a chi ld, all a reas — oral product ion and comprehens ion as wel l a s reading and wr i t ing— must be a s s e s s e d . Bas ing dec is ions concern ing a student 's schoo l p lacement on only one a rea of ability wou ld be shorts ighted. 39 4.21 SUMMARY Although this study found no relationships between fast mapping and oral or written second language skill, number of years exposure to a second language or overall academic proficiency, it is difficult to believe that fast mapping as it exists in researchers' minds has nothing to do with the acquisition of language the second time around. The explanation for the current findings would seem to lie elsewhere—in the effect of L2 experience on fast mapping or in the relative simplicity of the fast mapping task used in this study. Suggestions for further research include i) an examination of the fast mapping skills of monolingual children as compared to bilingual children of the same age and ii) an examination of school age children's fast mapping skills when presented with a task more complex than this study's. The task could be made more complex by a) using affective state words, b) presenting each word in a variety of contexts several times, c) using several new words at once and d) increasing the time lag between exposure and testing from 10 minutes to several days. It is possible that, using a more complex fast mapping task, the relationship between French Immersion success and fast mapping that was sought in this study would emerge. However, French Immersion has become a very complex issue and success in learning French is not the only criterion for being successful in the program. Given the high academic achievement implicitly expected of children in French Immersion, and the effects of home culture and the personality of the child on her/his performance, it is entirely possible that fast mapping will never be found to correlate strongly with success in French Immersion. Perhaps the search for any one predictor of such success is futile. Until such a predictor is found, educators and parents will need to use to the fullest extent both a range of measures of language proficiency and their knowledge of each child's motivation and temperament in order to decide who should remain in French Immersion. 40 R E F E R E N C E S A p e l , K., & K a m h i , A . G . (1987). Language impaired chi ldren's fast mapp ing sk i l ls : A muke is a muke [Summary] . P resen ted at the A S H A Annua l Convent ion , New Or leans . Barrett, M. (1978). Lex ica l deve lopment and overextens ion in chi ld language. Journa l of Ch i ld  L a n g u a g e , 5 ,205 -19 . B rown , H. (1987). Pr inc ip les of language learning and teaching (2nd ed.). N e w Je rsey : P r e n t i c e -Hal l . Bruck, M. (1978). T h e suitability of ear ly F rench immers ion programs for the l anguage -d i sab led chi ld. C a n a d i a n Journa l of Educat ion , 3(4), 51-72. Bruck, M. (1982). Language impaired chi ldren's per formance in an addit ive bi l ingual educat ion program. App l ied Psycho l ingu is t ics , 3 ,45 -60 . C a r e y , S . (1978). The chi ld as word learner. In M. Hal le, G . Mil ler & J . B resnan (Eds. ) , Linguist ic  theory and psycho log ica l reality, (pp.264-293). C a m b r i d g e , M A : MIT P r e s s . C a r e y , S . , & Bartlett, E. (1978). Acqui r ing a s ingle new word. P a p e r s and reports on chi ld language  deve lopment , 15 ,17-29 . Dept. of L inguist ics, Stanford Universi ty. Car ro l l , D. W . (1986). P s y c h o l o g y of language. Monterey, C A : B rooks /Co le . C la rk , E . V . (1983). Mean ing and concepts . In J . F lavel l & E. Ma rkman (Eds. ) , Handbook of Ch i ld  Psycho logy , Vo l III: Cogni t ive Development , (pp.787-840). N.Y.: J o h n Wi ley and S o n s . C u m m i n s , J . (1984). Bi l ingual ism and spec ia l educat ion: Issues in a s s e s s m e n t and pedagogy. Exeter : Shor t R u n P r e s s Ltd. D ick inson , D. K. (1984). First impress ions: Ch i ld ren 's knowledge of wo rds ga ined from a single exposure . App l ied Psycho l ingu is t ics , 5, 359-374. Do l l aghan , C . (1984). Fast mapping in normal and language impaired chi ldren. P roceed ings of the  S y m p o s i u m on R e s e a r c h in Ch i ld Language Disorders , 5 , 7 3 - 8 9 . Do l l aghan , C . (1985). Ch i ld meets word: Fas t mapping in p reschoo l chi ldren. Journa l of S p e e c h and  Hear ing R e s e a r c h , 28, 449-454. G e n e s e e , F. (1976). The suitability of immers ion programs for all chi ldren. C a n a d i a n Modern  L a n g u a g e Rev iew , 32, 494-515. G e n e s e e , F. (1978). Individual d i f ferences in s e c o n d language learning. C a n a d i a n Modern  L a n g u a g e Rev iew, 34, 490-504. 41 He ibeck , T. H. , & M a r k m a n , E. M. (1987). W o r d learning in chi ldren: A n examinat ion of fast mapp ing . Ch i ld D e v e l o p m e n t 58, 1021-1034. Lambert , W . E . , & Tucker , G . R. (1972). Bi l ingual educat ion of ch i ldren: T h e St. Lambert  exper iment . Row ley , M A : Newbury H o u s e Inc. Lambert , W . E . , Tucker , R. G . , & d 'Angle jan, A . (1973). Cogni t ive and attitudinal c o n s e q u e n c e s of bi l ingual schoo l ing : T h e St., Lambert project through G r a d e 5. Journa l of Educat ional  Psycho logy , 65(2), 141-159. M a c n a m a r a , J . (1982). N a m e s for things: A study of human learning. C a m b r i d g e , M A : MIT P ress . Ma ra t sos , M. P. , & Cha lk ley , M. A . (1980). T h e internal language of ch i ldren 's syntax. In K. E . Ne l son (Ed.) , Ch i ld ren 's Language , 2, 127-214. N.Y. : G a r d n e r P r e s s . Pe te rs , A . M. (1981). Language segmenta t ion : Operat ing pr incip les for the percept ion and analys is of language. In D. S lob in (Ed.), Un iversa ls of language acquis i t ion. Pe te rs , A . M. (1983). The units of language acquis i t ion. Cambr i dge , M A : C a m b r i d g e University P r e s s . R i c e , M. L . , Buhr , J . C . & Nemeth , M. (in press) . Fas t mapping word learning abil i t ies of language de layed p reschoo le rs . Journa l of S p e e c h and Hear ing R e s e a r c h . R i c e , M. L . , W o o d s m a l l , L. (1988). L e s s o n s from televis ion: Ch i ld ren 's word learning when viewing. Ch i l d Deve lopment , 59, 420-429. Ru thven , L. (1989). T h e use of cohes ive dev i ces by s c h o o l - a g e language impaired chi ldren. Unpub l i shed M .Sc . thes is , Universi ty of Brit ish C o l u m b i a . Shr ibe rg , L. D . , & Kent , R. D. (1982). C l in ica l phonet ics. N e w York : J o h n Wi ley and S o n s . S w a i n , M. (Ed.). (1972). Bi l ingual schoo l ing : S o m e exper iences in C a n a d a and the Uni ted States. Toronto: Ontar io Institute for S tud ies in Educat ion . S w a i n , M. , & Burnaby, B. (1976). Personal i ty character is t ics and s e c o n d language learning in young ch i ldren: A pilot study. Work ing papers in bi l ingual ism (O ISE) , IT, 115-128. Tr i tes, R. L. (1976). Ch i ld ren with learning difficulties in F rench immers ion. C a n a d i a n Modern  L a n g u a g e Rev iew , 33, 193-216. Tr i tes, R. L. , & Pr ice , M. A. (1979). A s s e s s m e n t of read iness for pr imary F rench immers ion:  K indergar ten fo l low-up assessmen t . Ot tawa: Universi ty of Ot tawa P r e s s . W a r d h a u g h , R. (1986). A n introduction to socio l inguist ics. Oxford , G B : Bas i l B lackwel l Ltd. 42 A P P E N D I X A T E A C H E R Q U E S T I O N N A I R E S tuden t , G r a d e Us ing the fol lowing rating sys tem: Leas t capab le pupi l in the ^ c l a s s Ave rage Mos t capab le 4 pupi l in the c lass 1.1 wou ld rate the student 's per formance on math ass ignments as fall ing within: a) R a n g e l b) R a n g e 2 c) R a n g e 3 d) R a n g e 4 2. I wou ld rate the student 's oral F rench ski l ls as fall ing within: a) R a n g e 1 b) R a n g e 2 c) R a n g e 3 d) R a n g e 4 

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:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0098089/manifest

Comment

Related Items