Verb agreement morphology in Hebrew-speaking children with specific language impairment.

Abstract

Earlier reports of verb morphology use by Hebrew-speaking children with specific language impairment (SLI) have suggested that these children mark agreement with the subject as accurately as younger control children matched according to mean length of utterance (MLU). This issue was examined in greater detail in the present study by including a wider range of agreement inflections from the present and past tense paradigms and employing verbs of different patterns (binyanim). It was hypothesized that children with SLI would be more limited than would MLU controls in their use of agreement inflections within past tense because the past tense agreement paradigm of Hebrew requires the simultaneous manipulation of three features-person, number, and gender. Differences between the groups were not expected for the use of agreement inflections within present tense, because only two features-number and gender-must be manipulated in the present tense paradigm. A group of preschool-age children with SLI was found to have more difficulty than did MLU controls in the use of most past tense agreement inflections. Within present tense, the two groups differed in their use of agreement inflections in only one pattern. For both groups, most errant productions differed from the target form by only one feature, usually person or tense. We found no feature that was consistently problematic for the children. The findings are discussed within a limited processing capacity framework.