Abstract
In this paper, we present the results of a new forced-choice task designed to test SLI children’s competence with three different agreement configurations: Determiner-Noun, Subject-Verb and Object-Verb agreement. Three populations of Italian-speaking children took part in the study and we compared the performance of a group of typically developing children with two groups of children diagnosed with phonological (P-SLI) or grammatical (G-SLI) Specific Language Impairment. Our study revealed that in this task the G-SLI group performed worse than the other two groups. We also found that the different agreement configurations under scrutiny introduced different degrees of complexity, with the Determiner-Noun condition being the easiest one. We discuss these results in relation to Clahsen’s (1997) Grammatical Agreement Deficit Hypothesis and to a more recent proposal presented in Moscati and Rizzi (2014). Furthermore, we also compared plural and singular S-V agreement morphology. Results indicate that in our comprehension task no extra cost is associated with plural morphology in none of the experimental groups.
