Participants were shown short video clips of body movements, without seeing faces or hearing sound, and were asked to identify the emotions conveyed by the body movement. Emotions were basic ones, such as anger, disgust, fear, happiness, and sadness. They were also given a motion coherence test: they were shown dots on a computer screen, with a certain number moving either left or right. The test had different percentages of dots moving coherently in one direction, creating different difficulty levels. There was a "strong link between performances on the tests" in this study, with neurotypical adults outscoring adults with ASD on both tasks.
The results suggest that adults with ASD have "difficulty in perceiving certain sorts of motion, particularly the movement of spatially separate elements spread over a relatively wide area that nevertheless move in the same direction, which is consistent with most previous findings."
This study is interesting because it specifically leaves out faces. Lots of studies have investigated whether people with ASD process faces differently, but there's a lot more to understanding other's mental states than just looking at their faces. A study like this gets at whether ASD problems are face-specific or affect processing of all kinds of social stimuli.
If visual processing is one of the contributing factors in at least some parts of the spectrum, that would be consistent with findings that sensory processing problems are particularly common in ASD kids. It would also be consistent with my guesses about why children with ASD tend to look away from faces (http://mosaicofminds.blogspot.com/2009/07/connection-between-looking-away-to.html#comments).
It'd also be interesting to look into this with kids...
I'd like to say more, but I'm still trying to find the original study. I have the citation to it:
Anthony Atkinson. Impaired recognition of emotions from body movements is associated with elevated motion coherence thresholds in autism spectrum disorders. Neuropsychologia, June 2009...but so far I'm not finding it, even at Durham University's website (where Anthony Atkinson is from). Will discuss more when I can see the actual results.