The phonology of two groups of SLI (n =5) and LD (n =5) children was analysed at age 3 and compared with two control groups: an age control (n =5) and a language level control (measured using the MLU-W) (n =5). Children with SLI and LD showed a delay in the acquisition of segments, syllabic structures and word structures, and in the simplification processes, compared with their age control group. However, SLI children also displayed significant differences vis-à-vis their language level controls, mainly in early acquisitions: vowels, nasals and stops at the segmental level, and in CV structures at the syllabic level. There is also a simplification process that seems to be more prevalent in these children than in their language level controls, namely, the deletion of unstressed syllables, mainly initial ones. The results enable SLI to be distinguished from LD and suggest that the development of SLI phonology is deviant. This deviation is interpreted as being a plateau in early acquisitions when later acquisitions have already appeared. The results are considered in the light of Leonard's surface hypothesis and an exclusively linguistic cause for this disorder is ruled out.

In this article we examine language processing and development in Catalan or Spanish-speaking children with SLI, focusing on the study of the verb. We analyse the key initial phase of its process of acquisition and aim to define common features of the SLI group that distinguish them from children with normal language development. We intend to identify more precisely the kind of delay shown by these children in a language with a rich verb morphology, in terms of both structure and chronology. The sample comprised 18 Catalan-Spanish bilingual pre-school children, assigned to three groups of six; an SLI group and two control groups, one matched for age and the other matched for MLU-w. Developmental data were obtained by recording situations of spontaneous speech at two different time points. Certain differences were found between groups in verb production. Production of verb inflection by children with SLI was only partial at the first evaluation; they maintained the same percentage of errors after a year. The patterns of correct and incorrect verb forms found in Catalan and Spanish do not corroborate the EOI hypothesis, but they support the Surface Hypothesis, given that the number of errors is not particularly high. This suggests the presence of limitations in subjects' processing ability, linked to the typological characteristics of the specific language being learnt.

Speech perception involves the integration of auditory and visual articulatory information, and thus requires the perception of temporal synchrony between this information. There is evidence that children with specific language impairment (SLI) have difficulty with auditory speech perception but it is not known if this is also true for the integration of auditory and visual speech. Twenty Spanish-speaking children with SLI, twenty typically developing age-matched Spanish-speaking children, and twenty Spanish-speaking children matched for MLU-w participated in an eye-tracking study to investigate the perception of audiovisual speech synchrony. Results revealed that children with typical language development perceived an audiovisual asynchrony of 666 ms regardless of whether the auditory or visual speech attribute led the other one. Children with SLI only detected the 666 ms asynchrony when the auditory component followed the visual component. None of the groups perceived an audiovisual asynchrony of 366 ms. These results suggest that the difficulty of speech processing by children with SLI would also involve difficulties in integrating auditory and visual aspects of speech perception.

This study investigates narrative comprehension and production in children with specific language impairment (SLI). Twelve children with SLI (mean age 5;8 years) and 12 typically developing children (mean age 5;6 years) participated in an eye-tracking experiment designed to investigate online narrative comprehension and production in Catalan- and Spanish-speaking children with SLI. The comprehension task involved the recording of eye movements during the visual exploration of successive scenes in a story, while listening to the associated narrative. With regard to production, the children were asked to retell the story, while once again looking at the scenes, as their eye movements were monitored. During narrative production, children with SLI look at the most semantically relevant areas of the scenes fewer times than their age-matched controls, but no differences were found in narrative comprehension. Moreover, the analyses of speech productions revealed that children with SLI retained less information and made more semantic and syntactic errors during retelling. Implications for theories that characterize SLI are discussed.