Auditory Global-Local Processing in Children with Autism Spectrum Disorders

Background:  

The human brain processes sensory information at different perceptual levels: at a “global” (i.e., whole) level, or at a “local” (i.e., detail) level.  Research from the visual domain suggests that individuals with autism spectrum disorder (ASD) have a more local-based processing style compared to typically-developing (TD) individuals (Plaisted et al., 1999).  Similarly, findings in audition show a local advantage in ASD.  However, auditory global processing in ASD is less clear, with studies showing either intact (Heaton, 2005) or impaired global processing (Foxton et al., 2003).  Moreover, the stimuli used in previous auditory global-local studies (called “interval-contour” stimuli) have been criticized for not measuring true global-local distinctions.

Objectives:  

The main objectives of the present research were: 1) to better characterize global-local processing distinctions in the auditory domain using new and improved auditory global-local stimuli; and 2) to test for group differences in auditory global-local processing between children with ASD versus TD. 

Methods:  

We are currently collecting auditory behavioral data and MRI measures in a large group of children with ASD versus TD controls as part of a multi-site study on brain and behavioral development (Zwaigenbaum et al., 2011).  Here we present preliminary data from 8 children with ASD (mean age 10.8, range 6-15 years; mean IQ 104, SD 12.4), and 9 TD children (mean age 10.7, range 7-13 years).

We used a new class of auditory global-local stimuli adapted from a study by Justus and List (2005).  These stimuli confer advantages over previous “interval-contour” stimuli in that they were designed to allow the independent manipulation of global and local structure (and are thus a better measure of true global-local distinctions), as well as direct comparison with analogous stimuli in the visual domain.  Stimuli consisted of 9-tone melodies, each comprised of three triplet (3-tone) sequences.  The global pattern was defined as the first tone of each triplet pattern, and the local pattern was defined as a single triplet.  Participants were asked to discriminate between ascending and descending pitch direction at the global and/or local level.  Ethical approval for this research was obtained by the Montreal Neurological Institute and Hospital Research Ethics Board.

Results:  

Preliminary findings in TD children showed a strong global advantage wherein global patterns were detected faster and more accurately compared to local patterns. In comparison, children with ASD showed a trend for a less pronounced global advantage, which appears to be driven by enhanced processing of local structure compared to TD.

Conclusions:  

We extend previous findings by demonstrating a less pronounced global advantage (due in part to enhanced local processing) in ASD.  However, we can conclude with greater confidence relative to previous work that global-local auditory processing differs in ASD versus TD.  Our preliminary findings are consistent with findings from studies in vision that used analogous visual global-local stimuli, suggesting that global-local processing is a perceptual phenomenon that is pervasive across sensory domains.  Results are consistent with current models of enhanced perception in ASD (Mottron et al., 2006).