Atypical visual processing as an endophenotype of autism spectrum disorders?

Background: Autism spectrum disorders (ASD) are highly heritable, but insight into the underlying etiology is still limited, mainly due to considerable heterogeneity within ASD. This large heterogeneity stimulates the search for more ‘genetically informative phenotypes’ or ‘endophenotypes’ that allow us to delineate more homogeneous subgroups. Endophenotypes are phenotypes that are more proximal to the biological etiology of a clinical disorder than its signs and symptoms, and are influenced by one or more of the same genes that confer susceptibility to the condition. Accordingly, it is postulated that a good ASD endophenotype should co-occur with ASD and should be expressed at a higher rate in unaffected first-degree relatives of ASD probands than in the general population.  Atypical visual processing has been suggested as a potential endophenotype for ASD. Enhanced local processing and reduced global processing abilities have been reported in ASD, as well as a more locally oriented processing style.

Objectives: The aim of this study is to assess which visual processing measures provide good endophenotype candidates for ASD.

Methods: In line with Booth (2006) four tasks were administered, targeting local versus global processing abilities at different levels (low- to mid-level versus higher-level visual tasks) and measuring processing style: a visual search task (local processing/low to mid-level), a coherent motion task (global processing/low to mid-level), a fragmented object outlines task (global processing/higher level) and the Rey-Osterrieth Complex Figure (processing style). Two reaction time tasks were also included to control for differences in motor speed. All tasks were administered from 315 individuals (IQ>70), covering five groups: children with ASD (n = 62), their first-degree relatives (unaffected siblings (n = 38) and parents (n = 91)), typically developing (TD) children (n = 63) and TD adults (n = 61). The age range in children and adults was 8-18 and 30-60 years, respectively.

Results: Analyses comparing children with ASD and TD children (matched for age, IQ and gender) showed a more locally oriented processing style and increased motion coherence thresholds in children with ASD. There were no significant group differences in visual search or motor speed. On the fragmented object outlines task, only subtle group differences were found, suggesting that participants with ASD needed slightly more information to correctly identify object contours. First-degree relatives of ASD probands did not perform differently from matched TD individuals.

Conclusions: Individuals with ASD showed a more locally oriented processing style, but no superior local processing ability, nor a general global processing deficit. Reduced global processing was only found on the coherent motion task, and only subtle differences were found on the fragmented object outlines task. Recent studies suggest that deficits on the coherent motion task may be due to a reduced ability to segregate signal from noise rather than reduced global integration. These findings, combined with the absence of group differences between ASD relatives and TD individuals indicate that visual processing abilities and visual processing style do not offer good endophenotypes for ASD.