International Meeting for Autism Research (London, May 15-17, 2008): NEURAL CORRELATES OF VERBAL AND NONVERBAL SEMANTIC INTEGRATION IN YOUNG CHILDREN WITH AUTISM

Friday, May 16, 2008

Champagne Terrace/Bordeaux (Novotel London West)

J. McCleery , DMC Laboratory of Cognitive Neuroscience, Harvard Medical School, Boston, MA

R. Ceponiene , Center for Research in Language, University of California, San Diego, La Jolla, CA

K. Burner , Center on Human Development and Disability, University of Washington, Seattle, WA

C. Williams , Center for Research in Language, University of California, San Diego, La Jolla, CA

M. Kinnear , Center for Research in Language, University of California, San Diego, La Jolla, CA

J. Townsend , Neurosciences, UC, San Diego, La Jolla, CA

L. Schreibman , Department of Psychology, University of California, San Diego, La Jolla, CA

Background: Previous behavioral and neuroimaging studies have shown that individuals with autism evidence impairments in semantic processing. However, it is not currently known whether these deficits are confined to the verbal domain or represent a more general problem with the processing of semantic information.

Objectives: To investigate verbal and nonverbal semantic integration in high functioning children with autism and typically developing children, aged 4 to 7 years, using high-density event-related potentials (ERPs).

Methods: The two groups were matched on chronological age, developmental age, and gender, and both groups scored in the normal range on standardized tests of language. Stimuli were matching and mismatching picture-word and picture-sound pairs. For example, a picture of a car was presented with (i) the word "car" (word match), (ii) the word "ball" (word mismatch), (iii) the sound of a car engine (sound match), or (iv) the sound of a bouncing ball (sound mismatch). N400 component effects were examined by subtracting matching word and sound ERPs from mismatching word and sound ERPs. Finally, t-tests were performed to determine whether N400 differences were different from zero.

Results: Children in both groups showed semantic mismatch effects in the environmental sound condition (CON Diff = 1.22 microvolts, s.e.=0.62, p<0.05; AUT Diff = 1.12 microvolts, s.e.=0.62, p<0.05). However, only controls showed differentiation of matching versus mismatching words (CON Diff = 1.27 microvolts, s.e.=0.54, p<0.05; AUT Diff = 0.01 microvolts, s.e.=0.55, p=0.49).

Conclusions: Because N400 match versus mismatch effects reflect the automatic detection of semantic mismatches, these results demonstrate that young children with autism exhibit abnormal automatic processing of semantic information at the level of single word meaning. The finding that match versus mismatch effects were normal in the environmental sound condition further suggests that this semantic processing abnormality may be limited to the verbal domain in these children.

See more of: Neurophysiology Posters
See more of: Poster Presentations