Michael Ullman and his colleagues at Georgetown University in Washington, D.C., are examining language learning and processing in autism, with a focus on grammar acquisition. Based on previous work examining grammar and its neural underpinnings, they predicted that children with autism would show abnormalities in grammar learning.

The researchers constructed an artificial grammar made up of syllables. This grammar can be learned over the course of a few minutes, allowing Ullman and his colleagues to examine the process of grammar acquisition in real time. In particular, the structure of the grammar allows the researchers to examine the hierarchical nature of grammar.

The researchers exposed children with autism and age-matched healthy controls to syllable strings generated by the grammar. The grammar-learning task was constructed in such a way that learning could be detected in real-time, as acquisition of the grammar ensued.

The data from this grammar-learning task has not yet been fully analyzed. However, related findings from a grammar task reveal faster-than-normal grammatical processing in children with autism versus typically developing children. This suggests that grammar acquisition and processing is indeed abnormal in autism, but in a way that may provide advantages to individuals with autism.