TD bilingual children

Comparing morphosyntactic abilities in a group of TD, Dyslexic and Bilingual Children Martina Caccia Neurocognition, Epistemology and Theoretical Syntax (NETS) Center of research, Institute for Advanced Study IUSS - Pavia, Italy

Background: Dyslexia, SLI & Bilingualism • SLI and Dyslexia are both developmental disorders: the first one affects the acquisition of oral language and it is often characterized in terms of problems with morphosyntax; the second one is a specific impairment in the acquisition of reading and spelling skills despite normal or above-average intelligence (Jakubowicz et al. (1998), International Dyslexia Association (1994)). • It is known that bilingual speakers, both children and adults, perform more poorly than monolinguals in some specific language domains, such as vocabulary, lexical access and morphosyntax (Bialystok 2008, 2010, Gollan and Kroll, 2001, Serratrice et al. 2004, Sorace 2011). Such lexical and morphosyntactic difficulties in bilingual children may bring to compare these children with those suffering from Specific Language Impairment (SLI) who typically show deficits in the lexical and in the morphosyntactic domain. However: • Vender et al. 2015 have shown that Early Second Language children present a linguistic profile which is qualitatively and quantitatively different from that typically shown by SLI children. This suggests that despite the superficial similarities it is possible to distinguish properly between the two populations.

The tests: Materials and Methodology

Results

Participants: 1) 2)

3) 4)

a control group (CG): 108 third and fourth graders (60 females), between 7;1 to 10;1 (M= 106; SD= 7,0). A group of children with developmental dyslexia (S)DD: 29 children, between suspected and diagnosed dyslexia, of third and fourth grade (17 males), age range 7;1-10;1 (M=105,7; SD= 7,7). A group of First Graders (FG) : 44 children (21 females and 23 males), between 5;7 - 6;9 (M=105,7; SD= 7,7). 22 TD bilingual children (BG) (M= 107,8; SD= 8,3): 9 Arabic, 1 Arabic/ French, 3 Spanish, 2 Chinese, 1 Filipino, 1 Russian, 1 English, 1 Albanian, 1 Croatian, 1 Romanian, 1

. Listening: (S)DD, CG and BG show about the some results. Differences arise with S-V agreement for FG (i.e.: *Gli abili pittori dipinge un quadro/*The skilled painters paints a picture) and with Cl/Pp for FG and BG.

1) Correct Answers After Listening/Reading correct sentences 120

100

80

. Reading: CG makes more errors than expected -> They tend to correct automatically, while reading, violations on D-N and S-V. S(DD) mistakes are probably due to their attempt to read as accurately as possible to the detriment of the accuracy of grammaticality judgment. BG shows about the same results in both modalities.

60

40

20

0 D-N

S-V S(DD) read

V-S

S(DD) listen

Cl/Pp

CG read

CG listen

S- Pp V. inac. FG

BG read

BG listen

French.

Agreement Test:

Two modalities: listening and reading. Task: grammaticality judgement and sentence’s correction. Materials: 26 sentences, 13 correct and 13 manipulated, through which 5 syntactic agreement configurations were assessed: 1) Det.- Noun (i.e. *Le mamma ha preparato la torta/(*The f,plur mum f,sing. has prepared the cake). 2) Subj. – Verb (i.e. *Il bambino hanno fatto i compiti/ (*The child have done his homework). 3) Verb – Subj. (i.e. *Esce le principesse/*Gets out the princesses). 4) Clitic – PastPart. (i.e. * La mamma li ha vestito/ (*The mum them m,plur. has dressed m.,sing). 5) Subj. – PastPart. Unaccusative Verb (i.e. *Le bambine sono caduta/ *The girls f.,plur.. are fallen down f.,sing).



Production of 3rd person clitic pronouns

(based on Prévost 2012, COST bi-SLI)

Some children have judged ungrammatical and they have corrected sentences without agreement manipulations, with Cl-Pp. (I.e. A sentence like: “Il ragazzo lo ha raccolto”/ (“*The boy it/him m.,sing has picked up f., sing” is mostly corrected by omitting the object clitic: “Il ragazzo ha raccolto”).

2) Wrong Answers After Listening/Reading correct sentences 50

40

S(DD) read CG read

30

S(DD) listen CG listen

20

FG BG read.

10

BG listen.

0 D-N

S-V

V-S

Cl/Pp

S- Pp V. inac.

3) Object Clitic Production 100

I.e.: Guarda! Qui c’è un gatto e qui un pesce. (Look! There are a cat and a fish.) Dimmi: Che cosa fa il gatto al pesce? (What does the cat to the fish?) • Clitic: Lo mangia. (Eat it) • DP: Mangia il pesce. (Eat the fish) • Null: Mangia. (Eat)

80

60

40

20

0 Clitic

DP S(DD)

CG

4)Individual Analysis of BG

Null FG

BG

In the light of some unexpected data, an individual analysis of each group was made: all answers related to syntactic structures with objects clitic were analysed. Such analysis has revealed some differences within BG (not only!): it was found a subgroup that behaves differently from other bilingual children. Obj. Cl. Prod. Refl.Cl. Prod. *Cl-Pp detect DG (Spanish) HR1 (Arabic) DV (Spanish) HR (Arabic) AM (Arabic)

8,3 0 8,3 50 16,6

60 0 100 80 0

100 66 100 33 100

Omission in correction *Cl-Pp

Cl-Pp detect.

Omis. In correct. Cl/Pp

0 66 100 0 50

33 0 0 66 100

67 100 100 34 0

Discussion: Object Clitic and Individual analysis The individual analysis has shown:  a subgroup (about 30%) in (S)DD group that shows difficulties with the object clitic both in the agreement test and in the production.  Subgroups in CG and in BG that show more difficulties in Cl-PastP agreement and in object clitic production than their peers. So…  Lack of production of clitic is a clinical marker of SLI in Italian and in other Romance languages at the age of 5 and, as data shows, it can persist beyond 5 years.  Even children with only diagnosis of Developmental Dyslexia, in third and fourth school-grade, can show severe problems in the production of clitics.  A fine-grained linguistic analysis can be really useful to identify possible cases of atypical development.  These data might well indicate that:  some child presents a latent SLI syndrome that can not be diagnosed using conventional tests.  it is importance to distinguish, in bilingual children, between superficial linguistic difficulties not related to SLI from actual clinical markers of the syndrome.

Selected references

Arosio F., Branchini C., Barbieri L., Guasti M. T., 2012. Evaluating morphosyntactic, semantic and pragmatic abilities in the search of persistent clinical markers of SLI in Italian children, CLAD project. Belletti, A. & Guasti ,M.T., 2015. The Acquisition of Italian. Different topics in different modes of acquisition, John Benjamins, chapters 1, 3, 7. Bishop D. V. M., Snowling M. J., 2004. Developmental Dyslexia and Specific Language Impairment: same or different?, Psychological bullettin, 130-6, 858-886. Bortolini U., Arfè B., Caselli M.C., De Gasperi L., Deevy P., Laurence, Leonard L.B., 2006. Clinical markers for specific language impairment in Italian: the contribution of clitics and non word repetition, International Journal of Language and Communication Disorders, 41(6): 695-712. Durrleman S., Delage H., Clitics across pathologies: Investigating the syntactic phenotypes of Autism, SLI and Developmental Dyslexia, presented at GALA, September 2013. Jakubowicz C., Nash L., Rigaut C, Gerard C.-L.,1998. Determiners and clitic pronouns in French speaking children with SLI, Language acquisition, 7, 2/4, 113-160. Moscati V., Rizzi L., 2012. The Selective Development of Agreement in Early Italian, STiL, Vol. 5, 91. Sorace A. 2011. Cognitive advantages in bilingualism: Is there a “bilingual paradox”? In P. Valore (ed.,) Multilingualism. Language, Power, and Knowledge, 335-358. Pisa: Edistudio. Vender, M. et al., Child L2 learning and Specific Language Impairment: superficially similar but linguistically different. More Than One Language in the Brain, 2015.

Contacts www.nets.iusspavia.it [email protected]