Background: The investigation of brain structure in autism has traditionally focused on volumetric differences by employing techniques such as voxel-based morphometry (VBM). VBM is, however, highly unspecific as observed differences might be attributed to several parameters related to cortical surface topography (e.g. cortical thickness or pattern of gyrification).

Objectives: The aim of this study was to identify which structural parameters in particular elicit the differences in brain anatomy previously observed in autism by employing both volume-based and surface-based topographic analysis.

Methods: Structural MRI scans were collected on 29 well-characterized individuals with autism (mean age = 32 yrs, mean IQ = 105, right-handed) and 27 age/gender matched healthy controls. A set of 4 measures (grey matter volume, cortical thickness, curvature index, folding index) were obtained for each anatomical region using FreeSurfer software, and subsequently compared across subject groups.

Results: Across all investigated areas, most differences were observed in curvature index and cortical folding including several areas of the frontal, parietal and temporal lobe, with individuals with autism displaying reduced parameter values. No significant group differences were observed in cortical thickness. A significant reduction in grey matter volume in the autistic group was observed in the left lateral orbitofrontal cortex and the bilateral rostral middle frontal area only.

Conclusions: The data shows that the previously observed differences in brain structure in the autistic brain are predominantly due to shape rather than volume abnormalities suggesting an abnormal pattern of regional neuronal connectivity. This finding is in agreement with a recent study demonstrating that individuals with autism display cortical folding abnormalities in parietal and frontal regions (Nordahl et al., 2007).