What do imaging studies tell us about the neural basis of autism?

There is no clear evidence from imaging studies for specific structural abnormalities in the brains of people with autism. The most robust observation is of greater total brain volume. There is evidence that this greater volume is not present at birth, but appears during the first few years. This brain enlargement might be a marker of abnormal connectivity due to lack of pruning. While abnormalities have often been reported in the cerebellum and the amygdala, these are difficult to interpret since both increases and decreases in the size of these structures have been observed. Another way of identifying the neural basis of autism is to investigate brain systems underlying cognitive functions compromised in this disorder such as face perception and 'theory of mind'. Autistic people fail to activate the 'fusiform face area' during face perception tasks and show weak activation of medial frontal cortex and superior temporal gyrus when performing theory of mind tasks. These problems stem from a lack of integration of sensory processing with cognitive evaluation. I speculate that this problem reflects a failure of top-down modulation of early sensory processing. The problem could result from abnormal connectivity and lack of pruning.