Cognitive Processing of Global and Local Visual Stimuli in High-Functioning Individuals with Autism

Background: Three decades ago researchers noticed that despite marked social and behavioral impairments, individuals with autism spectrum disorders (ASD) excelled on tasks requiring a focus on small parts or elements. Established deficit accounts of autism failed to explain these strengths. The weak central coherence (WCC) theory was developed to explain both impairments and superior skills observed in autism. This theory asserts that people with autism possess a weakness in the neural systems involved in pulling information together to establish meaning; this results in over focus on fine detail and disregard of context, or the global picture. The WCC theory has been supported by many studies, several groups, however, failed to find any marked deficits in global processing in people with autism. Consequently, a competing theory, enhanced perceptual functioning (EPF), was proposed. The EPF hypothesis asserts that people with autism have enhanced perceptual systems; this enhanced perception biases people with autism toward focusing on local details and parts. This bias, however, does not come at the expense of global processing. An ongoing debate is whether people with autism have an excessively detail-focused processing (local processing) style accompanied with impaired contextual processing (global processing). 

Objectives: The present study assesses both global and local processing, and shifts between global and local processing in an effort to reconcile ongoing controversy. To clarify cognitive processing style and perceptual ability in people with autism, this paradigm (1) tests several dimensions of the effects of local and global variables on processing, (2) employs both high- and low-level perceptual tasks, and (3) collects eye movement data.

Methods: Individuals with autism and age- (15-30 years) and IQ- (85+) matched normal controls were given an embedded figures task, two silhouette tasks, a local-global switching task, and a local and a global motion detection task. In the embedded figures task, participants located a target shape hidden in a larger complex figure (local processing advantageous). In the silhouette tasks, participants selected the correct silhouette of target images (global processing advantageous). In the local-global switching task participants identified rapidly presented local or global letters (both local and global processing required). In the motion detection tasks participants detected the motion of dot displays (global motion detection task) and vertical gratings (local motion detection task). Eye-tracking data was collected during the embedded figures task and silhouette tasks as an index of underlying strategy. 

Results: Overall the autism group exhibited superior performance on local processing, but abnormal performance on global processing. Eye-movement data (fixation duration and frequency) also suggests atypical perception underlies ASD performance.  

Conclusions: Our findings provide a more comprehensive understanding of perception in people with autism. The behavioral data helps clarify processing style in autism. The eye-movement data provides additional information about the perceptual and attentional strategies employed by individuals with autism. These findings combined help refine cognitive and neuroanatomical accounts of autism.