In the first year of life, language development is determined by two closely related factors: biological (the development of the speech organs) and cognitive-linguistic (perception and acquisition of the characteristics of the language heard in the environment) (Stoel-Gammon 2011: 2).
Comparative studies of language acquisition show that, up to 6 months of age, children of different nationalities babble in the same way: they first begin to produce plosives (p, t, k, p, d, g), nasals (m, n), and vowels (Velleman, Vihman 2007; Žukauskienė 2012). It is argued that children begin producing these sounds first not only because they are easy to articulate, but also because they hear them most often in their surroundings. For example, the articulation of v, z, and d or fricative sounds is more complex; they are also used less frequently (Velleman, Vihman 2007: 26; Edwards, Beckman 2008: 124). Somewhat later, these individual sounds begin to be combined into sound sequences with a CV (consonant-vowel) syllable structure (Velleman, Vihman 2007: 26), and CV sequences repeated several times resemble words in adult speech, e.g., mama, tete, baba, bamido, and so on (Stoel-Gammon 2011: 8).
From 6–7 months, children increasingly produce the sounds of their native language: constantly hearing child-directed speech1, they begin to acquire the sounds heard in their surroundings—
1 Child-directed speech is a register used by adults when communicating with children. It is characterized by a simplified vocabulary, short sentences of simple structure, and an abundance of nouns (especially diminutive forms) and verbs (Cameron-Faulkner et al. 2003; Kamandu-lytė 2006a).
E. KRIVICKAITĖ, I. DABAŠINSKIENĖ. Lietuvių kalbos fonologijos įsisavinimas: netikrų žodžių kartojimo eksperimentinis tyrimas
237
Page 2
sus, o kiti garsai išnyksta (Velleman, Vihman 2007: 26). Taip vyksta natūralus fonologijos įsisavinimas: vaikai, girdėdami aplinkos kalbą, pastebi dėsningumus ir pradeda jais remtis. Lyginamieji kalbų tyrimai atskleidė, kad, pvz., lūpiniai garsai, kurie yra būdingi prancūzų kalbai, dažniau girdimi 10 mėn. prancūzų vaikų čiauškėjimo laikotarpiu, nei švedų ar japonų kūdikių, kurių aplinkos kalboje šių garsų esama mažiau (Edwards, Beckman 2008: 123); švedų, estų ir bulgarų vaikai anksčiau išmoksta tarti v garsą nei anglakalbiai (Ingram 1988); l garsą prancūzų vaikai išmoksta tarti vėliau nei anglakalbiai (Chervrie-Muller, Lebreton 1973, cituojama iš Edwards, Beckman 2008: 3–4).
The influence of the language heard in the environment increases as the child’s vocabulary grows; over time, the child becomes able to pronounce increasingly complex and less frequent sounds and sound clusters (Velleman, Vihman 2007; Kavitskaya et al. 2011). It has been established that by around the age of three, Lithuanian children are already able to pronounce the following consonant clusters within a syllable: bl, dž, gt, kd, kl, km, kn, ks, kš, kt, lč, lg, ln, lp, nd, ng, nt, sk, sl, sp, st, sv, šč, tg, zd, žd and consonant clusters occurring between syllables: gs, gt, ks, kš, kt, lk, ls, lt, lv, ms, mt, nk, nt, ps, sk, sn, st, šd, šk, šm, šp, šs, št, tl, tn, tr, tv, ts, žd, žm, žs, žt. The most difficult clusters to learn to pronounce are kl and sp within syllables, and lv, nt, sk, st between syllables (Kamandulytė 2006b: 85–88). Somewhat later—around the age of four to five—children should already be able to pronounce clusters of two or three consonants well (although they may still omit a consonant) (Garšvienė, Ivoškuvienė 1993: 32–33). Children’s vocabularies become richer and their speech approaches that of adults: they use an increasing number and variety of derived words, their formation of compound words improves, and suffixed adjectives become more numerous (Garšvienė, Ivoškuvienė 1993: 37). Results from a non-standardized test of Lithuanian children’s sound pronunciation show that one fifth (78 percent) of preschool-aged children pronounce individual sounds correctly, while they most often make errors when pronouncing the sounds s, z, š, ž, c, dz, č, dž and r. A sound that has been learned is more difficult to pronounce within a word than in isolation or in a syllable (Garšvienė, Juškienė 2008: 80), so children make more errors when repeating words than when pronouncing individual sounds. Most often, sounds are replaced by other sounds, e.g., asilas → atilat, ropė → vopė, pjūklas → priūklas; sounds are also often omitted, particularly in consonant clusters, e.g., zebras → zebas, sviestas → siestas, herbas → hebas (Garšvienė, Juškienė 2008: 80–81). Thus, preschool-aged children still have difficulty pronouncing Lithuanian sounds that are articulated in complex ways and occur less frequently.
Attempts have been made to explain phonological acquisition—that is, the natural human ability to perceive a sound system and pronounce sounds and their combinations—through the theory of natural pho-
238
KALBOS KULTŪRA | 86
Page 3
nology2. The aim is to identify the difficulties children encounter in acquiring phonological features. It has been established that children find low vowels (a, e) easier to perceive than high vowels (u, i); voiced stops are more difficult to pronounce clearly than voiceless ones. Each sound or sequence of sounds can be classified according to how difficult it is to perceive and pronounce. Natural phonology is explained through the opposition: easy to pronounce vs. clear to perceive (Dziubalska-Kołaczyk 2004).
The phonological processes of each language are distinctive; that is, each language has its own characteristic sounds, sound combinations, and syllable and word structures (Dziubalska-Kołaczyk 2004), but people respond similarly to linguistic difficulties. To communicate effectively, a child must overcome physiological constraints. For example, English-speaking children need to learn and become accustomed to pronouncing consonant clusters, whereas children who speak Hawaiian do not, because consonant clusters are not characteristic of that language.
As already mentioned, language acquisition begins with the patterns most commonly used in a language: children first begin using the sounds and sound combinations they hear most often in their environment. Comparative studies show that children first acquire syllables with a CV structure, most often reduce consonant clusters, and do not pronounce final consonants within a syllable (Fikkert 2000). Researchers observe phonological processes occurring in children’s speech and most often mention distant assimilation and metathesis. It has been established that distant assimilation occurs in as many as 32 percent of the words used by children (Velleman, Vihman 2007: 33). Metathesis is also common in children’s speech: word-initial consonants are most often transposed with final ones.
It is thought that these processes occur in children’s speech because, when learning new words, children draw on their accumulated experience of word pronunciation. They pronounce their first words quite accurately (these are usually the first 10–20 words with uncomplicated structures), but pronounce later words less accurately. This phonological regression is characteristic of early speech, because children systematize the pronunciation patterns of their first words and rely on them when acquiring later words. In other words, children choose words that fit a pattern familiar from the babbling period (e.g., CVCV), so for words with more complex structures they try to find ways to make
2 The theory of natural phonology developed in the 20th century from the study of phonetics and phonology. Its founder was David Stampe (1969, 1973). Later, natural phonology developed under the influence of work in natural linguistics, especially studies associated with W. U. Dressler (1985, 1994, Dressler et al. 1987).
E. KRIVICKAITĖ, I. DABAŠINSKIENĖ. Lietuvių kalbos
fonologijos įsisavinimas: netikrų žodžių kartojimo eksperimentinis tyrimas
239
Page 4
pronunciation easier, e.g., by adding a vowel at the end of a word or simplifying the pronunciation of final consonants (Velleman, Vihman 2007).
Phonological development continues for a long time, so even preschool- or early school-aged children may have difficulty pronouncing more complex sounds, consonant clusters, or multisyllabic words. This process includes easier and more difficult cases, which are discussed in the following sections.
2. NONWORD REPETITION TEST
A great deal of useful information about later phonological development can be obtained by administering special tests. A nonword repetition test is a task in which participants are asked to repeat nonwords3. Repeating an unusual sound sequence, a meaningless word, is not easy—it requires multiple cognitive-linguistic abilities, including speech perception, phonological memory, articulation, and so on (Rispens, Parigger 2010). Studies show that children with language impairment find it particularly difficult to accurately repeat a meaningless word heard for the first time (Marshall, van der Lely 2009: 40). Findings from studies conducted abroad are consistent: children with language impairment perform statistically significantly worse on nonword repetition than their typically developing peers (Gathercole 2006a, 2006b; Rispens, Prigger 2010). For example, in a nonword repetition test administered to typically developing children and children with language-development impairment aged 8;3–10;1145, a statistically significant difference was found between the two groups, especially when comparing repetition of nonwords longer than two syllables (three to five syllables) (We ismer et al. 2011). It is therefore argued that the nonword repetition test can be used as a diagnostic tool to identify possible language-development impairment in both younger (4–5 years) and older (11–12 years) children (Stokes, Klee 2009).
In addition to its use in the diagnosis of early language development, nonword repetition tests are important for monitoring the development of vocabulary and other language levels. A child’s ability to repeat a new, multisyllabic word heard for the first time at an early age indicates readiness to learn new words and expand their vocabulary.
3 Nonwords are phonological sequences of sounds that conform to the phonotactic rules of a particular language but have no meaning and do not function as any part of a sentence (Clark et al. 2012: 578).
4 Age is denoted as follows: 8;3 means eight years and three months.
5 This paper discusses specific language impairment (English Specific Language Impairment – SLI). For more on this, see Ruzaitė, Dabašinskienė 2010.
240
KALBOS KULTŪRA | 86
Page 5
(Gathercole 2006a): poor repetition of a word heard for the first time may indicate difficulty learning new words or a language-development impairment. The nonword repetition test is associated with language acquisition: it is argued that nonword repetition begins in infancy, when children are just starting to learn their first words (Gathercole 2006a). In other words, every word a child once heard for the first time sounded unusual and new to them—just like the words they are asked to repeat when taking a nonword repetition test (Chiat, Roy 2007).
Among the first to present findings from a nonword repetition test were English psychologists S. E. Gathercole and A. Baddaley (1989): a longitudinal study of nonword repetition found a correlation between test results and subsequent vocabulary development in children. At the start of the study, the vocabulary size of 70 four-year-old children was measured, and they took a nonword repetition test. When the study was repeated a year later, the four-year-olds’ nonword repetition scores were found to correlate statistically significantly with the vocabulary size of the same children at age five: children who had achieved better results on the nonword repetition test a year earlier had larger vocabularies than those whose test results had been poorer (Baddeley et al. 1998: 160). Poorer nonword repetition test results may signal that a child will have difficulty learning and remembering new words.
Eksperimentinio tyrimo (Gathercole, Baddely 1990), analogiško netikrų žodžių kartojimo testui, rezultatai leido formuluoti dar vieną hipotezę apie vaiko žodyno raidą: kuo naujas žodis panašesnis į jau žinomą, tuo lengviau jį įsiminti. Šio tyrimo metu 5 m. amžiaus anglakalbių vaikų prašyta įsiminti įprastus vardus (Peter, Michael) ir neįprastus (Puemás, Meeton). Nustatyta koreliacija tarp šio testo ir netikrų žodžių kartojimo testo rezultatų: vaikai, kurių netikrų žodžių kartojimo testo rezultatai buvo prastesni, prasčiau įsiminė ir niekada negirdėtus vardus. Teigiama, kad netikrų žodžių pakartojimo tikslumas susijęs su fonologinės atminties gebėjimais: kuo naujas žodis panašesnis į jau žinomą ar kelia asociacijų su žinomu žodžiu, tuo lengviau jį išmokti (Gathercole, Baddely 1990).
The nonword repetition test also proved suitable for studying foreign-language learning. Finnish children aged 9–10 took a nonword repetition test based on the phonological features of English before beginning to learn English as a second foreign language. Two years later, when the English vocabulary size of the same children was assessed, a strong relationship was found between their nonword repetition test results from two years earlier and the size of the English vocabulary they had acquired over the intervening two years (Baddeley et al. 1998: 160).
E. KRIVICKAITĖ, I. DABAŠINSKIENĖ. Lietuvių kalbos
fonologijos įsisavinimas: netikrų žodžių kartojimo eksperimentinis tyrimas
241
Page 6
2.1. Main features of the nonword repetition test
Comparative studies of nonword repetition tests developed on the basis of the phonologies of different languages have identified two principal features of nonword structure that determine test results: 1) syllable structure (consonant–vowel (CV) or consonant cluster–vowel (CCV)); 2) word length (the number of syllables in a word) (Chiat, Roy 2007; Gutierrez-Clellen, Simon-Cereijido 2010). It is proposed that a nonword test should consist of those phonemes that children acquire earliest; in that case, repetition accuracy would not be affected by a physiological inability to produce a sound (Dollagham, Campbell 1998).
The results of nonword repetition tests in different languages around the world are consistent:
1) nonwords containing consonant clusters are harder to repeat than words without clusters (Santos et al. 2006; Marshall et al. 2009, Kapalkova et al. 2013; Kavitskaya et al. 2011);
2) consonant clusters occurring at the beginning of a word are repeated more accurately (Marshall et al. 2009: 42), because this is associated with louder and stronger production of the first syllable of the word (Pegy, Mork 2010). It has also been established that children first begin to produce consonant clusters at the beginning of words (around the age of two to three); consonant clusters are used more frequently and are more varied in this position (McLeod et al. 2001);
3) one- and two-syllable nonwords are repeated most accurately; that is, nonword repetition accuracy decreases as word length increases (Kavitskaya et al. 2011). When repeating nonwords, a child must rely on short-term phonological memory. Longer words take more time to perceive and repeat; their phonological representation gradually fades from short-term memory, and therefore repetition results are poorer for longer words. It is argued that poor short-term phonological memory function is associated with slow acquisition of individual words (Gathercole 2006a: 600–601).
Another important factor determining nonword repetition accuracy is the age of the participants: the older the child, the more accurately they repeat both shorter (one- to two-syllable) and longer (three- to five-syllable) nonwords (Santos et al. 2006; Kavitskaya et al. 2011; Kapalkova et al. 2013). Older children know more words made up of different combinations of sounds, which means they have more phonological knowledge and can repeat unfamiliar words heard for the first time more accurately (Munson et al. 2005: 1034).
242
KALBOS KULTŪRA | 86
Page 7
2.2. Structure of the Lithuanian nonword repetition test
Lithuanian nonwords6 were created with the structural features of Lithuanian words in mind: only sounds and syllable structures that are possible and most commonly found in Lithuanian words were used.
A Lithuanian word consists on average of 7 phonemes (most often around 5–8 phonemes) (Kazlauskienė 2010: 35–37), and two- to four-syllable words predominate (Karosienė, Girdenis 1990: 21). The most common syllable types in Lithuanian words are CV, CVC, CCV, CCVC (Karosienė, Girdenis 1994; Kazlauskienė, Raškinis 2008: 26–27), which means that a syllable in Lithuanian most often consists of 2–3 sounds (Karosienė, Girdenis 1994). It has been established that two thirds of the syllables in all Lithuanian words (about 65 percent) are open (CV, CCV, CCCV) (Girdenis, Karosienė 2010: 82); a syllable most often begins with the consonants k, l, s, t, p, n and ends in vowels: o, a, i, e, u (Girdenis, Karosienė 2004: 220); medial syllables in words contain many palatalized consonants, and front vowels predominate (Garšva, Girdenis 1996). The most common sounds making up a word are i, a, e, o and s, k, t, m, n, r (Karosienė, Girdenis 1993; Kazlauskienė, Raškinis 2009). The Lithuanian nonword repetition test includes words with only two-consonant clusters, since three-consonant clusters are too difficult for young children to pronounce. The most common two-consonant cluster types in Lithuanian are: RT7 (42 percent) (nt, rt, nk, rb); ST (18 percent) (st, šk, št, sp); TR (13 percent) (kl, pr, tr, tv, gr, bl, tl, pl); SR (6 percent) (sm, sn, šv, sl, sv, šl) (Girdenis 1995: 107–111; Kazlauskienė 2007: 37–40). Thus, based on these principal structural features of Lithuanian words and previous studies of the acquisition of phonological structure by Lithuanian children (Kamandulytė 2006b; Garšvienė, Juškienė 2008), Lithuanian nonwords were constructed (see Appendix).
The Lithuanian nonwords were designed to avoid resembling real Lithuanian words or evoking any associations with a familiar word, while still conforming to all the structural features of Lithuanian words: the consonant clusters in the nonwords consist of three types of two-consonant clusters: ST (39 percent), TR (33 percent), and SR (28 percent)8; word length ranges
6 Lietuviškas netikrų žodžių pakartojimo testas (aut. Dabašinskienė, Krivickaitė 2013) kurtas dalyvaujant COST ISO804 Language Impairment in a Multilingual society: Linguistics Patterns and the Road to Assessment projekte (2011–2013 m.).
7 S – fricative (s, z, š, ž, f, h, ch); T – stop (p, b, t, d, k, g); R – sonorant (l, m, n, r, v, j).
8 There are no RT-type consonant clusters in the nonword repetition test, because this cluster is possible only across syllable boundaries.
E. KRIVICKAITĖ, I. DABAŠINSKIENĖ. Lietuvių kalbos
fonologijos įsisavinimas: netikrų žodžių kartojimo eksperimentinis tyrimas
243
Page 8
from 2 to 4 syllables (from 4 to 9 phonemes), and the words’ syllables have CV, CCV, or CCVC syllable structures (see appendix).
The Lithuanian nonword test consists of 24 words divided into three groups: 8 words consisting of two syllables (4–6 phonemes), 8 of three syllables (6–7 phonemes), and 8 of four syllables (6–9 phonemes). Each group contains 2 words without consonant clusters and 6 words with consonant clusters: in three words, the consonant cluster is at the beginning of the word, and in three, in the middle. All consonant clusters occur within a syllable, except in two disyllabic words, where they occur both within a syllable and at a syllable boundary (see appendix).
The study’s authors wanted to test whether children can pronounce not only nonwords but also Lithuanian words with a similar structure; therefore, a matching Lithuanian word was selected for each nonword (see appendix).
The matches were selected with regard to:
1) word length: nonwords and real words have the same number of syllables, e.g., šve.la9 → sma.la (disyllabic), šku.li.nė → ska.re.lė (trisyllabic), ma.gvu.no.lė → re.kla.mė.lė (four-syllable);
2) syllable structure (the number of phonemes in a syllable) and consonant clusters, e.g., ke.mu → do.ja (two phonemes in each syllable); spa.de.ki → sta.tu.la (consonant clusters in the first syllables).
Other aspects (the articulatory features of consonants and vowels) were not taken into account, because identical words could not be found; however, an effort was made to keep the consonant clusters as similar as possible in articulation. For example, pr (bilabial plosive + alveolar apical approximant) was replaced by the match br (bilabial plosive + alveolar apical approximant): gap-rė → kobra; gu (velar plosive + labiodental approximant) was replaced by kl (velar plosive + dental apical approximant): magvuno-lė → reklamelė; šk (alveolar apical fricative + velar plosive) was replaced by sk (dental apical fricative + velar plosive): škulinė → skarelė; for several words, identical consonant matches were found between nonwords and real words: gitua → gatvė, klesta → klasta, sminto → smiltė (see appendix).
9 Syllables are separated by periods.
244
KALBOS KULTŪRA | 86
Page 9
3. A STUDY OF THE REPETITION OF LITHUANIAN NONWORDS AND REAL WORDS
Data from 25 children with typical language development were selected for this exploratory study10. The participants were divided into five age groups of five children each: 1) 4–5 years; 2) 5–6 years; 3) 6–7 years; 4) 7–8 years; and 5) 8–9 years. Data were collected in kindergartens and schools in Kaunas. Each child was tested individually: first, the Nonword Repetition Test11 was administered, followed by the Real Word Repetition Test.
The aim of this study was to compare the ability of children of different ages to repeat Lithuanian nonwords and real words. The hypothesis was that children would repeat real Lithuanian words more accurately than nonwords, but that all polysyllabic words and words containing consonant clusters would pose greater difficulties.
The results of the study allowed us to draw a preliminary conclusion consistent with findings from similar studies in other countries: children find nonwords—words they have never heard before—more difficult to repeat than real words. Children repeated Lithuanian nonwords with approximately 89% accuracy, compared with 95% accuracy for real words.
The study showed that two-syllable words (both real and nonwords) were repeated with equal accuracy (96%); however, differences emerged when comparing real and nonwords longer than two syllables: children repeated real three-syllable words just as accurately as two-syllable words (96%), while results were slightly poorer for four-syllable real words12 (93%) (see Fig. 2). The number of syllables has a greater effect on the accuracy of nonword repetition: two-syllable nonwords were repeated with 96% accuracy, three-syllable nonwords with 89%, and four-syllable nonwords with 82% accuracy (see Fig. 1).
A comparison of the results of the two tests shows that errors are most often made when repeating real and nonwords with the same structure—that is, words containing consonant clusters, e.g., smala and švela; kobra and gaprė; žoliapjovė and gosaklūni, etc.; however, errors occur more frequently when repeating nonwords (see Fig. 2).
10 Children with typical language development are considered to be those who have no language disorders and do not attend speech therapy.
11 The structure and administration methodology of the Lithuanian Nonword Repetition Test are described in greater detail in Krivickaitė 2013; Dabašinskienė, Krivickaitė 2013; Krivickaitė 2014.
12 Not all children knew the meanings of the real words either. For example, one participant said that they had never heard the word raskila and did not know what it meant. It may be assumed that this explains why the accuracy of repetition of the word raskila was not high. Clearly, many preschool- and primary-school-age children may not know a given less frequently used word or one with a more abstract meaning.
E. KRIVICKAITĖ, I. DABAŠINSKIENĖ. Lietuvių kalbos
fonologijos įsisavinimas: netikrų žodžių kartojimo eksperimentinis tyrimas
245
Page 10
FIG. 1. Results of real- and nonword-repetition tests by number of syllables in the word
The data presented in Figure 2 show that when repeating three- and four-syllable real words containing consonant clusters, errors are less frequent than when repeating nonwords: the longer the nonword containing a consonant cluster, the lower the repetition accuracy. For example, the three-syllable real word gyslotis was repeated with 96% accuracy, while the nonword pasuapi was repeated with 60% accuracy; the four-syllable real word svajotoja was repeated with 100% accuracy, while the nonword snaldina was repeated with 76% accuracy. The study shows that errors occur more often when repeating three-syllable nonwords with a consonant cluster in the middle of the word (repetition accuracy about 78%), whereas nonwords with an initial cluster were repeated with about 95% accuracy. The position of consonant clusters has little effect on repetition accuracy for four-syllable nonwords: words with a consonant cluster in the middle were repeated with about 80% accuracy, while those with an initial consonant cluster were repeated with about 84% accuracy. The position of consonant clusters is also not very important for repetition accuracy of real words; all words with consonant clusters were repeated with similar accuracy: three-syllable words with an initial consonant cluster, with 96% accuracy, and those with a medial cluster, with about 95% accuracy; four-syllable words with an initial consonant cluster were repeated with about 93% accuracy, and those with a medial cluster, with about 91% accuracy13 (see Fig. 2).
13 The results of the pilot Lithuanian Nonword Repetition test are consistent with findings from other countries: clusters at the beginning of a nonword are repeated more accurately than those in the middle of a word (Marshall, van der Lely 2009; Williams et al. 2013). However, a larger sample of participants would be needed to confirm this trend.
246
KALBOS KULTŪRA | 86
Page 11
FIG. 2. Repetition accuracy for real words and nonwords
Repetition results for real words and nonwords without consonant clusters are similar, e.g., šalikas – 100%, širuta – 96%, šokoladas – 100%, žadevina – 96%.
The results of the real- and nonword-repetition tests showed that all age groups repeated real words more accurately than nonwords, and the greatest difference was observed in the youngest participant group (ages 4–5; see Fig. 3): real words were repeated with 81% accuracy, while nonwords were repeated with only 58% accuracy. A greater difference between the results for real- and nonword repetition was also observed-
E. KRIVICKAITĖ, I. DABAŠINSKIENĖ. Lietuvių kalbos
fonologijos įsisavinimas: netikrų žodžių kartojimo eksperimentinis tyrimas
247
Page 12
FIG. 3. Repetition accuracy for real words and nonwords by age group
among children aged 7–814 (first-grade pupils). The test results for the remaining three participant groups (ages 5–6, 6–7, and 8–9) are similar: although the difference is not very pronounced, real words are repeated more accurately (see Fig. 3). All the data discussed are presented as frequency percentages, since the small sample does not allow statistically significant differences to be claimed; therefore, we regard the results of this pilot study as possible trends, which are plausible because they are supported by studies of early language development in other countries and in Lithuania.
Not only frequency trends, that is, quantitative characteristics, but also the ways in which familiar and unfamiliar words are articulated are important factors that indicate the state of language development in different age groups. When attempting to pronounce words with more complex structures, children seek ways to make pronunciation easier. The real- and nonword-repetition tests revealed several types of simplification common to all age groups: sound substitutions (31% of all errors15), consonant assimilation (8%), sound omissions (14%), and sound additions (8%).
14 Repetition accuracy for real words and nonwords improves in the older age groups; only the nonword results for the 7–8-year-old group are poorer. In studies of nonword tests in other countries, a statistically significant difference was found between preschoolers, first-graders, and second-graders. This is known as the school effect (Santos et al. 2006: 375). In first grade, children experience many changes (a new environment, friends, and educational system), so they may be somewhat unsettled, which may lead to poorer results on the nonword-repetition test.
15 Summed errors in nonword and real-word repetition. An error is defined as an inaccurately repeated nonword or real word.
248
KALBOS KULTŪRA | 86
Page 13
Sound substitution is one of the most common errors when repeating pseudowords. Children replaced consonants with l, n, t, e.g.: lamuvį → lasmulį, lasmutį, lasmunį; pasvapi → paslapi. The sound p was also often replaced with k or t, e.g.: maspulė → paskulė; plemuta → klemuta; nisparima → nistarima; spiratuša → starutuša. These substitutions were characteristic of all age groups except the oldest (8–9 years). When repeating real words, there was only one instance of sound substitution: žoliapjovė → žoliapjolė (in the 5–6-year-old group).
Several instances of distant assimilation were observed in the youngest group of participants (4–5 years): žadevina → žalevina; nisparima → nismarima. Stops (d, p) are assimilated by later-occurring sonorants (m, l). Distant consonant assimilation also occurred when repeating real words in the 4–5-year-old group: skanumėlis → skanulėlis; in abiliūpinis, m was assimilated to the preceding alveolar consonant l.
When repeating four-syllable pseudowords, children made more omission errors and more often omitted a syllable rather than an individual sound, e.g.: magunolė → maguno, magulė, maguolė; sulerité → sulitė; snalidina → sladina. Participants aged 4–5, 5–6, and 6–7 made omission errors. When repeating real disyllabic words, individual sounds were omitted (kobra → koba, klasta → kasta), whereas when repeating four-syllable words, syllables were omitted (žoliapjovė → žolia; skylamušis → skilamu). Real words were shortened by children aged 4–5 and 5–6.
This pilot study showed that children most often attempt to simplify words with more complex structures (by replacing some sounds with others or omitting particular sounds or syllables); however, several instances were observed in which children aged 4–5 added another consonant to a consonant (creating an additional consonant cluster), e.g.: pasvapi → pasvalpi; plemuta → plemouta; gatvė → gvatvė.
CONCLUSIONS
The results of the pilot tests involving repetition of real and pseudowords in Lithuanian confirm findings from studies conducted abroad: words never heard before and unfamiliar words are more difficult to remember and repeat than real words. The greatest difference between repetition of real and pseudowords emerged when repeating three- and four-syllable words: performance on pseudoword repetition was considerably poorer than on real-word repetition. Errors were most common when repeating real and pseudowords with the same structure, but when repeating pseudowords, children made mis-
E. KRIVICKAITĖ, I. DABAŠINSKIENĖ. Lietuvių kalbos
fonologijos įsisavinimas: netikrų žodžių kartojimo eksperimentinis tyrimas
249
Page 14
takes more often. It was found that pseudowords containing consonant clusters (especially in the middle of the word) are harder to repeat than words without consonant clusters. Children attempt to simplify words with more complex structures, shortening longer words or replacing some consonants with others. Errors are most common among children aged 4–5 and among children aged 7–8 who have just started school.
Children in all age groups had the greatest difficulty repeating multisyllabic pseudowords containing consonant clusters. The complexity of word structure requires greater effort to process information, more complex perceptual processes, and additional cognitive (memory) resources. Less frequently used, unusual elements of language are less natural and more difficult for young children to perceive.
LITERATŪRA
Baddeley A., Gathercole S. E., Papagno C. 1998: The Phonological Loop as a Language Learning Device. – Psychological Review 1, 158–173.
Cameron-Faulkner Th., Lieven E. Tomasello M. 2003: A construction based analysis of child directed speech. – Cognitive Science 27, 843–873.
Chiat Sh., Roy P. 2007: The Preschool Repetition Test: Am Evaluation of Performance in Typically Developing and Clinically Referred Children. – Journal of Speech, Language, and Hearing Research 50, 429–443.
Clark N. B., McRoberts G. W., van Dyke J. A., Shankweiler D. P., Braze D. 2012: Immediate memory for pseudowords and phonological awareness are associated in adults and pre-reading children. – Clinical Linguistics & Phonetics 26 (7), 577–596.
Dabašinskienė I., Krivickaitė E. 2013: Language delay, incomplete learning or SLI: a non-word repetition test for Lithuanian. Tarptautinis IX doikalbystės simpoziumas, 2013 m. liepos 15–17 d. Singapūras: Nanyango technologijos universitetas.
Dollaghan Ch., Campbell T. F. 1998: Nonword Repetition and Child Language Impairment. – Journal of Speech, Language, and Hearing Research 41, 1136–1146.
Dressler Wolfgang U. 1985: Morphonology. The dynamics of derivation, Ann Arbor: Karoma Press.
Dressler Wolfgang U. 1994: Evidence from first phases of morphology acquisition for linguistic theory: extragrammatic morphology and diminutives. – Acta linguistica hafniensia 27, 93–105.
Dressler Wolfgang U., Mayerthaler W., Panagl O., Wurzel Wolfgang U. 1987: Leitmotifs in Natural Morphology, Amsterdam: Benjamins.
250
KALBOS KULTŪRA | 86
Page 15
Dziubalska-Kołaczyk K. 2004: Modern Natural Phonology: the theory for the future. – English Language, Literature and Culture. Selected papers from the 13th PASE conference, Poznań. Prieiga per internetą: <http://www.google.lt/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&ved=0CDAQFjAA&url=http%3A%2F%2Fwa.amu.edu.pl%2Fkdk%2Fsites%2Fdefault%2Ffiles%2FModern%2520Natural%2520Phonology.doc&ei=wUxdUqKcK8Ojhg e6yIHoCg&usg=AFQjCNEwiYee16WPTemH2EeHsJeG6WrM-A&sig2=bKjbRJcopASfz9RnJfJfkg&bvm=bv.53899372,d.ZG4> (žiūrėta 2013 10 01).
Edwards J., Beckman M. E. 2008: Some cross-linguistic evidence for modulation of implicational universals by language-specific frequency effects in phonological development. – Language Learning and Development 4 (2), 122–156.
Fikkert P. 2000: Acquisition of Phonology. – The First Glot International State-of-the-Article Book. The Latest in Linguistics. Studies in Generative Grammar 48, Berlin/New York: Mouton de Gruyter, 221–250. Prieiga per internetą: <http://www.fikkert.com/fikkert_glot.pdf> (žiūrėta 2013 09 28).
Garšva K., Girdenis A. 1996: Bendrinės lietuvių kalbos fonologija ir morfonologija. – Kalbotyros darbai III tomas: 1988–2000, Vilnius: Mokslo ir enciklopedijų leidybos institutas, 2001, 200–242.
Garšvienė A., Ivoškuvienė R. 1993: Logopedija: Vadovėlis specialiosios pedagogikos fakulteto studentams, Kaunas: Šviesa.
Garšvienė A., Juškienė G. 2008: Priešmokyklinio amžiaus vaikų tarimo ypatumai. – Jaunųjų mokslininkų darbai 2 (18), 78–84.
Gathercole S. E., Baddeley A. D. 1989: Phonological memory deficits in language disordered children: Is there a causal connection? – Journal of Memory & Language 29, 336–360.
Gathercole S. E. 2006a: Nonword repetition and word learning: The nature of the relationship. – Applied Psycholinguistics 27, 513–543.
Gathercole S. E. 2006b: Complexities and constrains in nonword repetition and word learning. – Applied Linguistics 27, 599–613.
Girdenis A. 1995: Teoriniai fonologijos pagrindai, Vilnius: Mokslo ir enciklopedijų leidybos institutas.
Girdenis A., Karosienė V. 2010: Skiemens ir žodžio pirmųjų bei paskutinių fonemų dažnumas bendrinėje lietuvių kalboje. – Baltistica 2, 213–231.
Gutierrez-Clellen V. F., Simon-Careijido G. 2010: Using Nonword repetition Tasks for the Identification of Language Impairment in Spanish-English Speaking Children: Does the Language of Assessment Matter? – Learning Disabilities Research & Practice 25 (1), 48–58.
Ingram D. 1998: The acquisition of word-initial [v]. – Language and speech 31 (1), 77–85.
E. KRIVICKAITĖ, I. DABAŠINSKIENĖ. Lietuvių kalbos fonologijos įsisavinimas: netikrų žodžių kartojimo eksperimentinis tyrimas
251
Page 16
K a m a n d u l y t ė L. 2006a: Vaikiškosios kalbos ypatybės. – Kalbos kultūra 79, 264–273.
K a m a n d u l y t ė L. 2006b: Morfonotaktikos įsisavinimas. – Žmogus ir žodis 8 (1), 84–89.
K a p a l k o v a S., P o l i š e n s k a K., V i c e n o v a Z., 2013: Non-word repetition performance in Slovak-speaking children with and without SLI: novel scoring methods. – International Journal of Language, Communication Disorders 48 (1), 78–89.
K a r o s i e n ė V., G i r d e n i s A. 1990: Bendrinės kalbos žodžio ir skiemens statistinė struktūra. – Kalbotyra 1 (1), 36–48.
K a r o s i e n ė V., G i r d e n i s A. 1993: Bendrinės kalbos fonemų dažnumai. – Kalbotyra 2 (1), 28–38.
K a r o s i e n ė V., G i r d e n i s A. 1994: Lietuvių bendrinės kalbos skiemens tipų dažnumai. – Kalbotyra 3 (1), 34–42.
K a v i t s k a y a D., B a b y o n y s h e v M., W a l l s T., G r i g o r e n k o E. 2011: Investigating the effects of syllable complexity in Russian-speaking children with SLI. – Journal of Child Language 38, 979–998.
K a z l a u s k i e n ė A. 2007: Intervokaliniai priebalsiai: vienanarės ir dvinarės grupės. – Kalbų studijos 11, 36–42.
K a z l a u s k i e n ė A. 2010: Lietuvių kalbos žodžių foneminės struktūros dėsningumai. – Žmogus ir žodis 1, 35–41.
K a z l a u s k i e n ė A., R a š k i n i s G. 2009: Bendrinės lietuvių kalbos garsų dažnumas. – Respectus Philologicus 16 (20), 169–182.
K a z l a u s k i e n ė A., R a š k i n i s G. 2008: Lietuvių kalbos fonologinio skiemens struktūrinių modelių dažnumas. – Žmogus ir žodis 1, 24–31.
K r i v i c k a i t ė E. 2013: Išgalvotų žodžių testas: lietuvių vienakalbių ir dvikalbių vaikų kalbos tyrimas. Pranešimas tarptautinėje taikomosios kalbotyros konferencijoje Kalbos ir žmonės: erdvė, laikas, tapatybė. Vilnius: Vilniaus universitetas. Prieiga per internetą: <https://www.academia.edu/5762723/Isgalvotu_zodziu_testas_Lietuviu_vienakalbiu_ir_dvikalbiu_vaiku_kalbos_tyrimas>.
K r i v i c k a i t ė E. 2014: Netikrų žodžių kartojimo testas: lietuvių vienakalbių ir dvikalbių vaikų kalbos tyrimas. – Taikomoji kalbotyra 4, 1–22.
M a r s h a l l C. R., v a n d e r L e l y H. K. J. 2009: Effects of word position and stress on onset cluster production: evidence from typical development, specific language impairment, and dyslexia. – Language 85 (1), 39–57.
M c L e o d Sh., v a n D o o r n, R e e d A. V. 2001: Normal Acquisition of Consonant Clusters. – American Journal of Speech – Language Pathology 10 (2), 99–110.
M u n s o n B., K u r t z B. A., W i n d o r 2005: The Influence of Vocabulary Size, Phonotactic Probability, and Wordlikeness on Nonword Repetitions
252
KALBOS KULTŪRA | 86
Page 17
of Children With and Without Specific Language Impairment. – Journal of Speech, Language, and Hearing Research 48, 1033–1047.
P e g y M o r k P. K. 2010: Language-specific realizations of syllable structure and vowel-to-vowel coarticulation. – The Journal of the Acoustical Society of America 128, 1346–1356.
R i s p e n s J., P a r i g g e r E. 2010: Non-word repetition in Dutch-speaking children with specific language impairment with and without reading problems. – British Journal of Developmental Psychology 28, 177–188.
R u z a i t ė J., D a b a š i n s k i e n ė I. 2010: Specific Language Impairment: Adaptation of a Screening Test for Lithuanian. – Darbai ir dienos 54, 277–300.
S a n t o s F. H., B u e n o O. F. A., G a t h e r c o l e S. E. 2006: Errors in nonword repetition: bridging short- and long-term memory. – Brazilian Journal of Medical and Biological Research 39, 371–385.
S t a m p e D. 1969: The acquisition of phonetic representation. – CLS 5, 443–453.
S t a m p e D. 1973: A Dissertation on Natural Phonology, Bloomington: IULC.
S t o e l - G a m m o n C. 2011: Relationships between lexical and phonological development in young children. – Journal of Child Language 38, 1–34.
S t o k e S. F., K l e e Th., 2009: The Diagnostic Accuracy of a New Test of Early Nonword Repetition for Differentiating Late Talking and Typically Developing Children. – Journal of Speech, Language, and Hearing Research 52, 872–882.
V e l l e m a n S. L., M i h m a n M. V. 2007: Phonology development in infancy and early childhood: implications for theories of language learning. – Phonology in Context. (red. Pennington. M. C.), Luton: Palgrave Macmillan, 25–50.
Ž u k a u s k i e n ė R. 2012: Raidos psichologija: integruotas požiūris, Vilnius: Margi raštai.
W e i s m e r S. E., T o m b l i n J. B., Z h a n g X., B u c k w a l t e r P., C h y n o - w e t h J. G., J o n e s M., 2011: Nonword Repetition Performance in School-Age Children with and Without Language Impairment. – Journal of Speech, Language, and Hearing Research 43, 865–878.
W i l l i a m s D., P a y n e H., M a r s h a l l Ch., 2013: Non-word Repetition Impairment in Autism and Specific Language Impairment: Evidence of Distinct Underlying Cognitive Causes. – Journal of Autism and Developmental Disorders 43, 404–417.
Gauta 2013 10 20
Priimta 2013 12 20
E. KRIVICKAITĖ, I. DABAŠINSKIENĖ. Lietuvių kalbos
fonologijos įsisavinimas: netikrų žodžių kartojimo eksperimentinis tyrimas
253
Page 18
PRIEDAS. Lietuvių kalbos netikri ir tikri žodžiai
Skiemenu sk.
Netikras žodis
Tikras žodis
Netikrų žodžių priebalsių samplaika → tikrų žodžių priebalsių samplaika
2
Kemu
Tema
–
2
Doja
Koja
–
2
Skimo
Stalas
Sk → st
2
Švela
Smala
Šv → sm
2
Gaprė
Kobra
Pr → br
2
Gitva
Gatvė
Tv → tv
2
Sminto
Smiltė
Sm → sm; nt → lt16
2
Klesta
Klasta
Kl → kl; st → st17
3
Geloša
Karosas
–
3
Širuta
Šalikas
–
3
Škulinė
Skarelė
Šk → sk
3
Plemuta
Klimatas
Pl → kl
3
Maspulė
Raskila
Sp → sk
3
Lasmuvi
Vyšnelė
Sm → šn
3
Spadeki
Statula
Sp → st
3
Pasvapi
Gyslotis
Sv → sl
4
Sulertė
Serenada
–
4
Žadevina
Šokoladas
–
4
Snalidina
Svajotoja
Sn → sv
4
Spiratuša
Skylamušis
Sp → sk
4
Nisparima
Restoranas
Sp → st
4
Magvunolė
Reklamėlė
Gv → kl
4
Staligosa
Skanumėlis
St → sk
4
Gosakluni
Žoliapjovė
Kl → pj
PASTABA. Pajuodinta priebalsių samplaika; ketvirtoje lentelės skiltyje pažymėta, kokia tikrame žodyje esanti samplaika atitinka netikrame žodyje esančią priebalsių samplaiką.
16 Netikrame žodyje sminto priebalsių samplaika nt yra tarp skiemenų, tikrame žodyje smiltė priebalsių samplaika lt taip pat tarp skiemenų.
17 Netikrame žodyje klesta priebalsių samplaika st yra tarp skiemenų, tikrame žodyje klasta priebalsių samplaika st taip pat tarp skiemenų.
254
KALBOS KULTŪRA | 86
Page 19
ACQUISITION OF LITHUANIAN PHONOLOGY: NONWORD REPETITION EXPERIMENTAL STUDY
Summary
A child’s capacity to repeat a new polysyllable word heard for the first time in the early age shows the ability to memorise new words and to expand one’s dictionary (Gathercole 2006a); therefore, the nonword repetition test is associated with language acquisition. It is stated that the repetition of nonwords starts very early in life when a child starts learning the first words (Gathercole 2006a): each word once heard by a child for the first time sounded unusual and new – like the words which are asked to be repeated in the nonword repetition test (Chiat, Roy 2007).
The article presents a pilot experimental study carried out on the basis of nonword and word repetition tests. The Lithuanian Nonword Repetition Test was designed with due regard to the structural characteristics of Lithuanian words (word length, vowel and consonant frequency, syllable structure); the test consists of 24 nonwords of different structure. Each nonword is associated with a Lithuanian equivalent taking into account word length and syllable structure. The sample of the pilot study is made of 25 typically developing preschool and school-age children (from 4 to 9 years old).
The results of the study revealed that the accuracy of repetition of nonwords and words is determined by three factors:
1) syllable structure: most errors occur in the repetition of words and nonwords containing combinations of consonants. What is more, the accuracy of repetition of nonwords is also determined by the place where the consonant combination occurs: more errors occur in repeating words with a consonant combination in the middle of the word compared to those with a consonant combination in the beginning of the word. The place of consonant combination does not affect the accuracy of repetition of words;
2) word length: the longer the word, the lower the accuracy of its repetition;
3) age of participants in the study: the older the age group, the more accurately both shorter (one-two syllable) and longer (three-four syllable) words and nonwords are repeated.
