Audiovisual crossmodal correspondences in autism spectrum disorders (ASDs)

The label ‘crossmodal correspondences’ has been used to define the nonarbitrary associations that appear to exist between different basic physical stimulus attributes in different sensory modalities. For instance, it has been consistently shown in the neurotypical population that higher pitched sounds are more frequently matched with visual patterns which are brighter, smaller, and sharper than those associated to lower pitched sounds. Some evidence suggests that patients with ASDs tend not to show this crossmodal preferential association pattern (e.g., curvilinear shapes and labial/lingual consonants vs. rectilinear shapes and plosive consonants). In the present study, we compared the performance of children with ASDs (6‐15 years) and matched neurotypical controls in a non-verbal crossmodal correspondence task. The participants were asked to indicate which of two bouncing visual patterns was making a centrally located sound. In intermixed trials, the visual patterns varied in either size, surface brightness, or shape, whereas the sound varied in pitch. The results showed that, whereas the neurotypical controls reliably matched the higher pitched sound to a smaller and brighter visual pattern, the performance of participants with ASDs was at chance level. In the condition where the visual patterns differed in shape, no inter-group difference was observed. Children’s matching performance cannot be attributed to intensity matching or difficulties in understanding the instructions, which were controlled. These data suggest that the tendency to associate congruent visual and auditory features vary as a function of the presence of ASDs, possibly pointing to poorer capabilities to integrate auditory and visual inputs in this population.