Cognitive component of the Componential Model of Reading (CMR) in Different Orthographies

R. M. JoshiTexas A & M University

mjoshi@tamu.edu

• COST Action, Prague, Czech RepublicCOST Action, Prague, Czech Republic

• November 7, 2015

• P. G. Aaron, Indiana State University, USA

• Zvia Breznitz, University of Haifa, Israel

• Florina Erbeli, University of Ljubljana, Slovenia

• Elena Grigorenko, Yale University, USA, and Moscow State University, Moscow, Russia

• Pooja Nakamura, American Inst. of Research, USA

• S. Nishanimath, Samveda Research Center, India

• Blanca Quiroz, Stanford University, USA

• Tao Sha, Cognitive Neuroscience Lab, Beijing, China

• Nandini Singh, National Brain Research Centre, India

••

•1. Introduction

•2. Incidence and Consequences of Reading Problems

•3. Reasons for Illiteracy

•4. CMR - Applications

•5. CMR – Various Orthographies

•6. Conclusions

• (2012) about 780 million people in the world are (2012) about 780 million people in the world are illiteratesilliterates

•Same as the entire population of the continent Same as the entire population of the continent of Europeof Europe

•120 million in Europe may have difficulty with 120 million in Europe may have difficulty with reading and writingreading and writing

•http://www.soschildrensvillages.org.uk/news/inthttp://www.soschildrensvillages.org.uk/news/international-literacy-day/illiteracy-the-facts-and-ernational-literacy-day/illiteracy-the-facts-and-figuresfigures

• Approximately 60 million adults and school-age children in the U.S. have difficulty acquiring reading and spelling skills.

• 1/3rd of fourth grade students • Among minority, inner-city children about 2/3rd (67%) • 25% of adults are unable to read a newspaper• Illiteracy affects all parts of life• (NIH) Illiteracy is a public health issue.••

• Health care expenses (USA)Health care expenses (USA)

• 33rdrd grade and below: grade and below:

• 10,700 dollars / year10,700 dollars / year

• at least 4th grade: at least 4th grade:

• 2,900 dollars / year2,900 dollars / year

• about 3.5 timesabout 3.5 times

• (Weiss et al., 2004)(Weiss et al., 2004)

• Illiteracy costs > 1 trillion USD each year Illiteracy costs > 1 trillion USD each year (report of the WLF, 2015) (report of the WLF, 2015)

•>50% of the adolescents with criminal problems and history of substance abuse have reading problems

•≈ 15% drop out of high school

•over 75% report difficulties in learning to read

•Number of prison cells

•“Based on this year's fourth-grade reading Scores, California is already planning the number of new prison cells it will need in the next ten years.”

Paul Schwartz, U. S. DoE

••

••

•Reasons for illiteracy:• Instructional and Environmental Reasons (Vellutino

& Scanlon, 2003)

•A. Instructional Reasons:

•Ehri (1989): Inadequate instruction spawning limited reading and spelling development and limited phonological awareness is the primary cause of reading disability

•Blachman, Texas group, Florida Group, . .••••

Poor instruction resulting in poor reading performance is especially true at the early primary grades.

•Juel (1988): Children who read poorly at the end of the first grade were likely to remain poor readers at the end of the fourth grade.

•Lyon et al. (1993): 74% of reading disabled in the third grade continue to exhibit reading and spelling problems even at the ninth grade level.

••

• B. Environmental Reasons:

•Oral language Development (Hart & Risley, 1995)

•Welfare families: 10 million → 500

•Middle class families: 20 million → 700

•Professional families: 30 million → 1100

•Number of books available at home (Chiu & McBride-Chang, 2006)

•Parents reading to children (Feitelson, 1964)

• Enjoyment of reading (Chiu & McBride-Chang, 2006)

•Genetics (Colorado Family Reading study, 1973-)

•Gender Differences (Chiu & McBride-Chang, 2006)

•

Das et al, 2010, 2011

English

ChineseTan et al., 2004

Brain organization - The universal reading network

Das et al, Neuroimage, 2011

Hindi

•Not only oral language but written language •Seymour et al. examined the speed and accuracy of

familiar word reading and nonword reading in 13 orthographies

•Finnish, German, Spanish, & Greek 95 %

•Portuguese, French, and Danish ≈ 75%

English = 34 % for word and 29% for non-word reading.

Brains may be similar from one culture to another but Brains may be similar from one culture to another but orthographies certainly are not.orthographies certainly are not.(Coltheart, Marshall, Newcombe)(Coltheart, Marshall, Newcombe)

Componential Model of Reading

Componential Model of Reading

Domain III

Ecological Components

Word recognitionComprehension

Domain I Domain II

Cognitive Components

Psychological Components

Motivation & InterestTeacher

ExpectationsGender Differences

Learned Helplessness

Home EnvironmentParental Involvement

Classroom Environment

DialectOrthography

Componential Model of Reading

Simple View of Reading

•R = D X C

•If D = 0, then R = 0;

•If C = 0, then also R = 0

•Joshi & Aaron (2000)

•R = D X C + F

•

The role of orthography in literacy acquisitionThe role of orthography in literacy acquisition

Orthographic DepthOrthographic Depth

Shallow……………………............…………………....…DeepShallow……………………............…………………....…Deep

Simple Finnish GreekItalianSpanish

Portuguese French

Complex GermanNorwegianIcelandic

DutchSwedish

Danish English

Syllabic Structure Source. Seymour, Aro, and Erskine (2003).

● English French Czech German Spanish

No. of letters ● 26 ● 26 39(13v + 26C)

● 30 ● 28-29 (w)

No. of phonemes

44 (20V + 24C)

●

38(19V + 19C)

37(6V + 25C)

≈30 29 (5V + 17C)

● Phoneme - letter ratio

1.7:1 1.5:1 1:1 1:1 1:1

● No. of graphemes

≈ 250 ≈165 42 ≈30 29

Japanese KanaJapanese Kana

•

•@@

•&&

• ಅ ಆ ಇ ಈ ಉ ಊಋಎ ಏ ಐ ಒ ಓ ಔ ಅಅ ಅಅ

• ಕ ಕಕ ಕ ಕಕ ಕಕ ಕಕ ಕಕ ಕಕ ಕಕಕ ಕಕಕ ಕಕಕ ಕಕಕಕ ಕಕ ಕಅ ಕಅ (Ka, Ki, Ke, Ko, . . .• ಖ ಖಕ ಖ ಖಕ ಖಕ ಖಕ ಖಕ ಖಕ ಖಕಕ ಖಕಕ ಖಕಕ ಖಕಕಕ ಖಕ ಖಅ ಖಅ• ಗ ಗಕ ಗ ಗಕ ಗಕ ಗಕ ಗಕ ಗಕ ಗಕಕ ಗಕಕ ಗಕಕ ಗಕಕಕ ಗಕ ಗಅ ಗಅ• ಘ ಘಕ ಘ ಘಕ ಘಕ ಘಕ ಘಕ ಘಕ ಘಕಕ ಘಕಕ ಘಕಕ ಘಕಕಕ ಘಕ ಘಅ ಘಅ• ಙ ಙಕ ಙಙ ಙಙಕ ಙಕ ಙಕ ಙಕ ಙಕ ಙಕಕ ಙಕಕ ಙಕಕ ಙಕಕಕ ಙಕ ಙಅ ಙಅ• ಚ ಚಕ ಚ ಚಕ ಚಕ ಚಕ ಚಕ ಚಕ ಚಕಕ ಚಕಕ ಚಕಕ ಚಕಕಕ ಚಕ ಚಅ ಚಅ• ಛ ಛಕ ಛ ಛಕ ಛಕ ಛಕ ಛಕ ಛಕ ಛಕಕ ಛಕಕ ಛಕಕ ಛಕಕಕ ಛಕ ಛಅ ಛಅ• ಜ ಜಕ ಜ ಜಕ ಜಕ ಜಕ ಜಕ ಜಕ ಜಕಕ ಜಕಕ ಜಕಕ ಜಕಕಕ ಜಕ ಜಅ ಜಅ• ಝಝಕಝಝಕಝಕಝಕಝಕಝಝಕಝಕಝಝಕಝಕಝಅಝಅ• ಞಞಕ ಞಙ ಞಙಕ ಞಕ ಞಕ ಞಕ ಞಕ ಞಕಕ ಞಕಕ ಞಕಕ ಞಕಕಕ ಞಕ ಞಅ ಞಅ• ತ ತಕ ತ ತಕ ತಕ ತಕ ತಕ ತಕ ತಕಕ ತಕಕ ತಕಕ ತಕಕಕ ತಕ ತಅ ತಅ• ಥ ಥಕ ಥ ಥಕ ಥಕ ಥಕ ಥಕ ಥಕ ಥಕಕ ಥಕಕ ಥಕಕ ಥಕಕಕ ಥಕ ಥಅ ಥಅ• ದ ದಕ ದ ದಕ ದಕ ದಕ ದಕ ದಕ ದಕಕ ದಕಕ ದಕಕ ದಕಕಕ ದಕ ದಅ ದಅ• ಧ ಧಕ ಧ ಧಕ ಧಕ ಧಕ ಧಕ ಧಕ ಧಕಕ ಧಕಕ ಧಕಕ ಧಕಕಕ ಧಕ ಧಅ ಧಅ• ನ ನಕ ನ ನಕ ನಕ ನಕ ನಕ ನಕ ನಕಕ ನಕಕ ನಕಕ ನಕಕಕ ನಕ ನಅ ನಅ• ಟ ಟಕ ಟ ಟಕ ಟಕ ಟಕ ಟಕ ಟಕ ಟಕಕ ಟಕಕ ಟಕಕ ಟಕಕಕ ಟಕ ಟಅ ಟಅ• ಠ ಠಕ ಠ ಠಕ ಠಕ ಠಕ ಠಕ ಠಕ ಠಕಕ ಠಕಕ ಠಕಕ ಠಕಕಕ ಠಕ ಠಅ ಠಅ• ಡ ಡಕ ಡ ಡಕ ಡಕ ಡಕ ಡಕ ಡಕ ಡಕಕ ಡಕಕ ಡಕಕ ಡಕಕಕ ಡಕ ಡಅ ಡಅ• ಢ ಢಕ ಢ ಢಕ ಢಕ ಢಕ ಢಕ ಢಕ ಢಕಕ ಢಕಕ ಢಕಕ ಢಕಕಕ ಢಕ ಢಅ ಢಅ• ಣ ಣಕ ಣ ಣಕ ಣಕ ಣಕ ಣಕ ಣಕ ಣಕಕ ಣಕಕ ಣಕಕ ಣಕಕಕ ಣಕ ಣಅ ಣಅ• ಪ ಪಕ ಪ ಪಕ ಪಪ ಪಪ ಪಕ ಪಕ ಪಕಕ ಪಕಕ ಪಪಕ ಪಕ ಪಅ ಪಅ• ಫ ಫಕ ಫ ಫಕ ಫಪ ಫಪ ಫಕ ಫಕ ಫಕಕ ಫಕಕ ಫಫಕ ಫಕ ಫಅ ಫಅ• ಬ ಬಕ ಬ ಬಕ ಬಕ ಬಕ ಬಕ ಬಕ ಬಕಕ ಬಕಕ ಬಕಕ ಬಕಕಕ ಬಕ ಬಅ ಬಅ• ಭ ಭಕ ಭ ಭಕ ಭಕ ಭಕ ಭಕ ಭಕ ಭಕಕ ಭಕಕ ಭಕಕ ಭಕಕಕ ಭಕ ಭಅ ಭಅ• ಮಮಕ ಮಮಕ ಮಕ ಮಕ ಮಕ ಮಮಕ ಮಕಮಮಕ ಮಕ ಮಅ ಮಅ• ಯಯಕಯಯಕಯಕಯಕಯಕ ಯಯಕಯಕಯಯಕಯಕಯಅಯಅ

• ರ ರಕ ರ ರಕ ರಕ ರಕ ರಕ ರಕ ರಕಕ ರಕಕ ರಕಕ ರಕಕಕ ರಕ ರಅ ರಅ (ra, ri, re, ro, . . . .• ಲ ಲಕ ಲ ಲಕ ಲಕ ಲಕ ಲಕ ಲಕ ಲಕಕ ಲಕಕ ಲಕಕ ಲಕಕಕ ಲಕ ಲಅ ಲಅ• ವ ವಕ ವ ವಕ ವಪ ವಪ ವಕ ವಕ ವಕಕ ವಕಕ ವವಕ ವಕ ವಅ ವಅ• ಶ ಶಕ ಶ ಶಕ ಶಕ ಶಕ ಶಕ ಶಕ ಶಕಕ ಶಕಕ ಶಕಕ ಶಕಕಕ ಶಕ ಶಅ ಶಅ• ಷ ಷಕ ಷ ಷಕ ಷಕ ಷಕ ಷಕ ಷಕ ಷಕಕ ಷಕಕ ಷಕಕ ಷಕಕಕ ಷಕ ಷಅ ಷಅ• ಸ ಸಕ ಸ ಸಕ ಸಕ ಸಕ ಸಕ ಸಕ ಸಕಕ ಸಕಕ ಸಕಕ ಸಕಕಕ ಸಕ ಸಅ ಸಅ• ಹ ಹಕ ಹ ಹಕ ಹಕ ಹಕ ಹಕ ಹಕ ಹಕಕ ಹಕಕ ಹಕಕ ಹಕಕಕ ಹಕ ಹಅ ಹಅ• ಳ ಳಕ ಳ ಳಕ ಳಕ ಳಕ ಳಕ ಳಕ ಳಕಕ ಳಕಕ ಳಕಕ ಳಕಕಕ ಳಕ ಳಅ ಳಅ

•Simple View of Reading:•Gough & Tunmer (1986)• Hoover & Gough (1990)

••English – Spanish bilinguals •Grades 1-4 50-60%

● Language ● Grade levels ● Total variance explained

● Results

● Enblish● Tilstra et al. (2009)

● 2-10 ● 40-70%; ● Grade 4 = 61%● Grade 7 = 48%● Grade 9 = 38%

● Decoding: 4th graders = 42%; 7th graders = 13%

● L C: 4th graders = 19%; 7th graders = 35%

● French● (Megherbi, Seigneuric, & Ehrlich 2006)

● 1 & 2 ● >50% ● Grade 1: Decoding = 27%; LC = 39% (10% shared)

● Grade 2 : Decoding = 16%; LC = 44% (8% shared)

● Norwegian

● Høien-Tengesdal & Hoien (2012)

● 6 ● 49% ● Mostly explained by LC; minimal contribution from decoding to RC from age 9

● Swedish● Høien-Tengesdal & Hoien (2012

● 6 ● 50% ● Mostly explained by LC; minimal contribution from decoding to RC from age 9

● Dutch: de Jong and van der Leij (2002)

● Greek: Protopapas, Sideridis, Mouzaki & Simos, (2007)

● Italian: Tobia & Bonifacci (2015)

● 1-3 ● 50% ● LC contributed much of the variance after grade 1

• Application of CMR to other languages• 1. Spanish, Chinese, Hebrew (Joshi et al., 2010, 2012, 2015)

•Chinese:•Grade 2; Character recognition & LC = 25%•Grade 4; Character recognition & LC = 42%•Character Rec. Grade 2 = 22% ; Grade 4 = 32%•Listening Comp. Grade 2 = 11%; Grade 4 = 31%•Hebrew•37% (Grade 6) to 70% (Grade 4) •D Grades 2 = 27%; 4 = 26%; 5 = 20%; 6 = 8%•LC Grades 2 =17%; 3 = 26%; 9 = 60%•

• ESL •First study of SVR (Hoover & Gough) was based on

bilingual/ESL population•Geva & Farnia (2012)•Longitudinal study grades 2-5•ELL and EL1 showed similar trend (explaining more than

60% of the variance)•Decoding more important in early grade levels•However, in EL1 LC contributed more to RC earlier and

ELL struggled with language tasks

•Verhoeven & van Leeuwe; Dutch as a second language•••

• Nakamura, Joshi, de HoopNakamura, Joshi, de Hoop•

• N = 556• Grades 2-5 • Schools from urban ‘slum’ communities and rural villages• Multilingual • Biliteracy in Primary Literacy (Lit1) Kannada/Telugu; and

Secondary Literacy (Lit2) English • Mother Tongues: Kannada (N= 78); Telugu (N=132);

Marathi (N=6); Tamil (N=45); Hindi (N=3); Urdu (N=10)••

● Low Elementary●

● High Elementary●

● B ● SE B ● β ● B ● SE B ● β

● PA ● .11 ● .12 ● .13 ● -.11 ● .10 ● -.11

● Dec ● .69 ● .17 ● .56*** ● .68 ● .14 ● .53***

● LC ● .08 ● .21 ● .08 ● .41 ● .15 ● .28**

● R2 = 45% ● R2 = 49%

Multiple Regressions by Grade

Note. PA = Phonological Awareness; Dec = Decoding; LC = Language Comprehension; RC = Reading Comprehension; * p < .05; ** p < .01; ***p < .001

•Approximately 50% of the variance was explained by the two factors

•Decoding in ‘akshara’ plays a stronger role even at the fifth grade level.

•L2 much better after a threshold (0.6) is reached in L1

•EFL (Erbeli & Joshi) Slovenia

•N = 480 seventh graders (271 = skilled) (209 = LS)

•Even though 60% of the variance was explained by two factors, LC was a better predictor of RC for skilled readers

•Decoding for less skilled readers.•

•CMR applied to bilinguals

•

•

• Dyslexia: Poor decoding and good comprehension• Hyperlexia: Good decoding and poor comprehension• Nation (1999), “Although poor comprehenders show less

severe deficits, they are similar to hyperlexic children in terms of the pattern of strengths and weaknesses of their reading skills” (p. 347).

• Gregorenko, Klin, and Volkmar (2003): “Some researchers are adamant about hyperlexia being a clinical phenomenon whereas others readily assign the label of hyperlexia to children with word-recognition-comprehension discrepancies irrespective of any clinical diagnosis” (p. 1080).

•Dyslexia •2,470 •2/3rd - English•Monolinguals•Bilingual Alexia,

(Dejerine, 1892; Hinshelwood, 1902)

•Pringle Morgan, 1896; 14 year old boy

••

•

•Hyperlexia •22•Almost all-English•Monolinguals•None from

neuropsychological patients

•Silberberg & Silberberg (1967)

•Dyslexia in bilinguals• Hinshelwood (1895) • 58 year old teacher of French and German languages• Hinshelwood (1902), ‘ . . how is it that there are so few

recorded cases of these partial forms of word-blindness, that is, cases of dissociation in polyglots? I think the reason is simply that the patient is not thoroughly examined by testing his power of reading all the characters and all the languages with which he is familiar

• Obler (1984): Unfortunately, there is virtually no literature on Obler (1984): Unfortunately, there is virtually no literature on childhood dyslexia in bilinguals.childhood dyslexia in bilinguals.

• Klein and Doctor (2003) studied 3 cases of biscriptal dyslexics Klein and Doctor (2003) studied 3 cases of biscriptal dyslexics of English and Afrikaans.of English and Afrikaans.

• Problems in both the languagesProblems in both the languages• Abu-Rabia & Siegel (2002)Abu-Rabia & Siegel (2002)• Arabic-English bilinguals in CanadaArabic-English bilinguals in Canada• Poor in Arabic were also poor in English tasks; bilingual poor Poor in Arabic were also poor in English tasks; bilingual poor

readers performed better on certain tasks (non word; spelling) readers performed better on certain tasks (non word; spelling) than monolingual English poor readersthan monolingual English poor readers

•Abu-Rabia & Siegel (2003)•Less skilled readers were poor in phonological

ability in Arabic, Hebrew and English.•Wydell & Butterworth (1999)•16 year-old English/Japanese bilingual boy•Problem only in English but not in Japanese•

• McBride-Chang, Liu, Wong, Wong, & Shu (2014)

•PC, PE, & PB: poor in PA tasks; •PC & PB: poor in MA•PB: RAN•Psycholinguistic Grain Size theory (Ziegler &

Goswami, 2005): differences in strategy during the reading acquisition process arise from the size of the speech unit represented by each written unit in a script.

••

•MS and VN: 16 years •Comparison (8): 3-10; 3-16; and

2-14 years old. •Background:

Tests administered• Raven’s Progressive Matrices (RPM)

• Letter/character naming: English & Kannada

• Decoding (nonword and real word): English & Kannada

• Listening Comp. : Word level (synonym judgment; grave-tomb) and passage level

• Reading Comp. : passage and cloze formats

• Spelling (dictation)

• Speed: letters/words

• PA

• (Joshi, et al., Dyslexia, 2010)

Tests RPM (avrg. = 47)

Letter namingEnglishKannadaNonword reading

EnglishKannadaWord reading(Coltheart’s list)Regular words

MS (16 yrs.)43

26/2658/58

27/2815/15

37/39 (Barge-barg; Strewn-Struwen)

VN (16 yrs.) 50

26/2658/58

20/289/15

32/39 (county-country; check-click; shrug-shur; barge-braj; spear-spar)

Comp.47

(45-50)

26/2658/58

28/2815/15

36/39(35-37)

TestsWord reading(Coltheart’s list)Irregular words

Joshi-Aaron list

Kannada

MS (16 yrs.)

31/39 (mainly regularization errors,

pint, lose-loss)

51/56 (Bacon-bakaan; ghost-aghast; tomb-tomb; sew-sue; lose-loss)

147/150

VN (16 yrs.)

17/39 (cough-cot; glove-goal; subtle-sublet; debt-dealt)

42/56 (past-post; beak-bake; tomb-mob; wool-owl; thorough-through)

131/150

Comp.

37/39 (pint)(36-38)

54/56 (tomb; sew)(53-55)

148/150(147-149)

TestsListening Comprehension

(Word level)Synonym judgment (grave-tomb)

EnglishKannada

Passage Comp.(English & Kannada)

MS (16 yrs.)

51/7664/80

25%

VN (16 yrs.)

62/7679/80

95%

Comp.

76/7680/80

100%

Tests

Reading Comp.(passage)EnglishKannada

Reading Comp. (Cloze format)

EnglishKannada

MS (16 yrs.)

25%30%

7/19 (37%)4/18 (22%)

VN (16 yrs.)

70%70%

12/19 (63%)10/18 (56%)

Comp.

100%100%

19/1918/18

TestsSpeed(letters; 40)

English

Kannada

Words (150)Wpm

Total time

Total errors

MS (16 yrs.)

17.30

14.99

93-1 = 92

110 Secs

3/150

VN (16 yrs.)

23.22

21.22

37-2 = 35

500 Secs

19/150

Comp.

17.85(16.5-19)

16.16(14-17)

80 (79-81)

130 secs(125-135)

2/150

TestsPhonologicalAwareness

Rhyme Rec.Syllable deletionSyllable reversalPhoneme oddityPhoneme deletionPhoneme reversal

MS (16 yrs.)

E K12/12 12/12

12/12 12/12

12/12 12/12

12/12

24/24

12/12

VN (16 yrs.)

E K12/12 12/12

12/12 12/12

10/12 11/12

9/12

12/24

2/12

COMP.

E K12/12 12/12

12/12 12/12

12/12 12/12

11/12

24/24

12/12

Sample spelling errors  MS (Hyperlexic) VN (Dyslexic) Comp.break – brake sign – sing break - brakethief – theif thief – thife sew - sowhygiene – hygene thorough-thoughtdeath – deadth wool-woon socks-seack

soup-spupe•

• ‘Face’ MS (2 mins): • tongue, appear, nourished, community, child,

logic, direct, brain, skull, graduate, eucalyptus, tamarind, rotten, zoology, botney, astrology, throat, widow, pupil, and orthodux.

• Others: wrote parts of the face - eye, ear, • or some adjectives- cheerful, blush

Performance of MS & VN in English and Kannada

Key: WR = Word Reading, LC WL = Listening Comprehension Word Level, LC PL = Listening Comprehension Passage Level, RC PL = Reading Comprehension Passage Level, RC CF = Reading Comprehension Cloze Format, SP = Spelling

•Conclusions•1. Illiteracy is of societal and national concern;

affects all walks of life•2. Decoding (D) and linguistic comprehension (LC)

can explain much of the variance in reading comprehension (RC) while IQ scores predict about 25% of the variance in RC.

•3. Decoding contributes more at the early grade levels and comprehension more at the upper grade levels. Decoding may play an important role in reading comprehension for a more prolonged time in a more opaque orthography. Language comprehension becomes more important for reading comprehension from the beginning to the more advanced stage.

1.1.

2.2.

•4. There are qualitative and quantitative differences in decoding and comprehension between hyperlexics (MS) and dyslexics(VN).

•5. Literacy acquisition and literacy problems among bilinguals may be influenced by the type of writing systems and the ‘orthographic distance’ between the two languages.

•6. Considering that the majority of the world’s population is bilinguals, more research studies on bilinguals are needed

••

•• Thank you Děkuji Mnohokrát

• Danke schön Спасибо

• Muchisimas Gracias בבה רר בדה תו•Bedankt Merci beaucoup ευχαριστώ

• धन्यववाद 谢谢 لل شڪ جـزي لا ر

Tack så mycketTusen takk

•

•