Auditory-visual synchrony could explain why kids with ASD prefer mouths to eyes

Ever wondered why sensory integration problems often go along with ASD? It may not be a coincidence.

Specifically, a recent study* suggests that sensory integration problems could cause young children to look at mouths rather than eyes, which in turn causes them to miss important social cues and fall behind in social development.

The study found that 2 year olds with ASD are strongly attracted to an exact match between motion and sound. The researchers started out with a standard design often used in research on face processing. They showed kids right-side-up and reversed, upside-down cartoons. Children and adults respond differently to inverted faces, which engage different brain circuits. An actor's voice, recorded when the animations were made, accompanied the presentations. The 2 animations were presented side by side, and researchers tracked the direction of children's gaze to determine which they preferred (based on looking time). Neurotypical and developmentally disabled kids without ASD not surprisingly preferred the upright animations, but toddlers with ASD showed no preference for the upright cartoons.

They did, however, strongly prefer the right-side-up animation in one case. The cartoon showed an upright figure clapping its hands during a game of pat-a-cake. The movement of the cartoon and the sound of the clapping were only in sync on the right-side-up animation. So the researchers redid their study and stimuli to see if they could find a consistent preference for visual-auditory synchrony in the ASD toddlers--and they did.
"Audio-visual synchronies accounted for about 90 percent of the preferred viewing patterns of toddlers with ASD and none of unaffected toddlers," said Jones. "Typically-developing children focused instead on the most socially relevant information."
Lip motion synchronizes with speech sound, while eye motion does not. Thus, children with ASD have a very good reason to be drawn to mouths instead of eyes.

Notice that this explanation does not propose that kids don't respond to or can't process eyes, just that they are much more drawn to mouths. This could explain why kids with ASD will look at eyes when cued to do so--there's no damage to eye viewing, just a preference for looking at mouths, which can be overcome at least temporarily through training.

Why would auditory-visual synchrony matter so much to the ASD kids? Perhaps they have trouble syncing up auditory and visual input on their own, so they seek that out in their environment. Neurotypical kids can sync up auditory and visual input on their own, allowing them to pay attention to social cues.

In other words: what might be going on here is an auditory-visual integration problem. We could test this hypothesis by repeating the experiment, only this time comparing ASD kids and neurotypical ones to kids with sensory integration disorders.

Integration per se wouldn't have to be the problem, though. Integration might make it easier for ASD kids to use one sense to make up for another. Most likely, they use auditory information to disambiguate visual input. We've seen a study suggesting that kids with ASD have trouble perceiving a gesture as a coherent motion. Importantly, kids were shown video clips without accompanying sound. Kids were then unable to use sound to sort out what they were seeing.

It's not surprising that kids with ASD look at bodies more than faces. People make noise as they move, and their movements are synchronized with this noise. There's a lot more opportunities for synchrony in the whole body than there is in the mouth alone. This holds true whether kids with ASD have an integration problem or a visual problem, but it'd happen differently. If they have an integration problem, they're seeing the gesture and hearing the sound at the same time, forcing them to integrate. If they have a visual problem, the sound allows them to see the gesture.

*Klin A, Lin DJ, Gorrindo P, Ramsay G, Jones W. Two-year-olds with autism fail to orient toward human biological motion but attend instead to non-social physical contingencies. Nature, Mar 29, 2009