We have examined gyral folding in a total of more than 500 subjects with first episode schizophrenia, subjects at high risk who do and do not become ill, people with learning disabilities (LD) with and without schizophrenia, and LD with schizotypal or autistic features, as well as appropriate healthy controls.

The gyrification index (GI), the ratio of the inner and outer cortical surface contours, was hand-traced bilaterally on every second 1.88-mm image slice throughout the brain in about 100 scans. We then developed an Automated-GI (A-GI) approach to determine cortical folding in pre-frontal lobes, and have applied this to the other scans.

Gyrification index values were significantly increased in the right temporal lobe of the schizophrenic patients. Right prefrontal lobe GI values were significantly increased in high risk individuals who subsequently developed schizophrenia (especially in BA 9 and 10). A-GI reduces the analysis time, improves repeatability, has low susceptibility to scanner noise and variability. Using A-GI we have replicated hand-traced results and also found a similar pattern of increased ‘gyrification’ in LD with schizophrenia or schizotypy but not LD alone or with autistic features.

Differences in fronto-temporal GI might reflect trait disconnectivity predictive of schizophrenia across a range of IQ levels. GI is however poorly understood and influenced by age, sex and volume measures. Further examination of sulco-gyral patterns is required to clarify this. A-GI could be usefully applied to MRI data sets of the brain in health and disease to address these issues.