Familial language impairment: more English evidence

Subjects between the ages of 9 and 22 with a history of familial language impairment were tested on a wide range of linguistic tasks. The data show that they are not different from the non-impaired members of their family or from external controls in their ability to hear and understand the morphological marker for plural or to understand a narrative, but they are significantly different in recognizing or correcting grammatical errors, in producing tense or number marking, and in understanding sentences with non-canonical word order. This pattern of deficits is similar to that which occurs cross-linguistically and in on-line processing tasks. These data support the hypothesis that these subjects do not build normal representations in their grammar nor construct productive grammatical rules. However, they are able to use other means such as memory and explicit rules to simulate grammar.