The perceptual and verbal analogical reasoning in children with specific language impairment (SLI)

1_{Department of Cognitive Sciences, University of Liège, Belgium} ² INSERM-Modyco, University of Paris X Nanterre, France

Sandrine Leroy1_{, Mélanie Guénébaud}1_{, Christophe Parisse², & Christelle Maillart}1

Address for correspondence: Sandrine Leroy, Department of Cognitive Sciences, University of Liège, Rue de l’Aunaie, 30, B38, 4000 Liège, BELGIUM, E-mail: Sandrine.Leroy@ulg.ac.be

References:

Bybee, J. (2010). Language, Usage and Cognition. United Kingdom: Cambridge University Press.

Markman, A. B., & Gentner, D. (1993). Structural alignment during similarity comparisons. Cognitive Psychology,

25(4), 431-467.

Thibaut, J.-P., French, R., & Vezneva, M. (2010). The development of analogy making in children: Cognitive load and executive functions. Journal of Experimental Child Psychology, 106(1), 1-19.

I

NTRODUCTION

G

OAL AND PREDICTIONS

Goal

To test performance of children with SLI to:

- alinguistic(syllables) analogical reasoning task

- anon linguistic(pictures) analogical reasoning task

Predictions

Children with SLI would preferentially use perceptual

similarityrather than relational similarity to solve analogical

reasoning tasks

 Children with SLI would have more difficulties to detect relational similarity between sequences when perceptual cues decrease

The perceptual and verbal analogical reasoning

in children with specific language impairment (SLI)

P

T

Children have tocomplete a test sequenceby choosing which picture among three others "goes best" in comparison with thetwo reference

sequences.

The4 possible solutionsare linked to a touch of the keyboard.

Placement of atyping mask(to avoid the child pressing an undesired key)

Illustration of the typing mask (non linguistic task) Illustration of the typing mask (inguistic task)

R

ESULTS

Usage-based theories(Bybee, 2010)consideranalogical reasoningas a cognitive process underlyingconstruction of more abstract linguistic schemas.

When children hear two utterances such as “John eats an apple” and “She sees a bird”, they can infer the abstract schema [S V O] by realizing analogies between the two utterances. According to Gentner and Markman (1997): "Analogy occurs when comparisons exhibit ahigh degree of relational

similaritywith verylittle attribute similarity". So, we distinguish perceptual similarity (easier but less important) from relational similarity.

If children with specific language impairment (SLI) havedifficulties to detect relational similaritybetween forms, we predictedmorphosyntactic

disorders, what would hinder their abstraction of construction schemas. Consequently, children with SLI would beless productivewith their language

and would use morefixed linguistic forms. Consequently, their morphosyntactic development would be slow down.

20 children with SLI

• Aged from 8 to 11

• Monolingual French speakers • QIP (WISC IV) > 82

• Language skills below 1.25 SD from the mean in 2 or more of 5 language areas • No neurological or auditory disorders

20 children with NL

• No history of language disorders • Monolingual French speakers • Matched with children with SLI

based on: (1)Chronological age; (2) QI performance; (3) Sex; and (4)

Socioeconomic background

D

ISCUSSION

Illustration of the non linguistic task Illustration of the linguistic task Four variables:

Group: SLI vs NL

Modality: linguistic vs non linguistic Inter-sequences similarity: with vs without Intra-sequence similarity: two vs one vs no common feature

No significantinteraction effect between: Modality - Group

F(1,36) = 3.4, p > .05

Inter-sequences similarity - Group

F(1,36) = 1.4, p > .05

Significantinteraction effect between: Intra-sequence similarity - Group

F(2,72) = 10.010, p < .001 SLI NL 2 common features 1 common feature no common feature

Intra-sequence similarity

D

ISCUSSION

No significant difference of modality: difficultiesnot specific to language

Children with SLI more dependent on perceptual cues, as well in non linguistic task as in linguistic task

Problem with detection of relational similaritieswithout perceptual cues

Generalization from exemplars without perceptual similarity could be

difficult, what could explain a slowed down morphosyntactic development

Role plays by working memory and inhibitory control in analogical reasoning(Thibaut, French & Vezneva, 2010)

Po s te r p re s e n te d at th e C L S in N e w C a s tl e , En g la n d , J u n e 1 3 –1 4 , 2 0 1 1