16693
The Neural Correlates of Perceptual Closure in Adults and Elderly with Autism

Friday, May 16, 2014
Atrium Ballroom (Marriott Marquis Atlanta)
P. C. M. Koolschijn and H. M. Geurts, Dutch Autism & ADHD Research Center, Brain & Cognition, University of Amsterdam, Amsterdam, Netherlands
Background:
Perceptual closure refers to the ability to form a global and coherent perceptual representation on the basis of few details. A classical example is provided by two-tone (black and white) images of human faces (Mooney faces). In a Mooney image, the local features become too ambiguous to be recognized individually, and must be disambiguated based on their context within a global configuration. Mooney faces have been used to investigate various aspects of intact and impaired face processing. Given that autism is typically associated with a detail-oriented visual processing style, the neural mechanisms underlying perceptual closure may be different in patients with autism spectrum disorders (ASD).

Objectives:
Examine the neural mechanisms underlying perceptual closure in adults and elderly with ASD.

Methods:
Using 3T event-related fMRI we measured BOLD-signal changes in 27 patients with ASD (31-76yrs) to investigate how the brain forms perceptual decisions about complex visual forms. One hundred upright faces and 100 visually similar non-face images were shown for 200ms in a random sequence. To identify closure-related activity contrasts of correctly identified faces minus non-faces were computed. Post-scanning tests included a self-paced shortened version of the same perceptual closure task to examine the overall ability to recognize face stimuli. To assess global-local interference processing, a Navon hierarchical figures test (large “global” letters composed of smaller “local” letters) was administered.

Results:
Behavior
Preliminary data suggest that participants with ASD show significant reduced detection rates (M=65% vs 92%; p<.001), but not reaction times (p>.1) during the perception of Mooney faces compared to non-faces. Performance inside the scanner and the self-paced task outside the scanner was significantly different, such that performance on the self-paced task showed higher hit-rates for faces (M=79%; p<.01) and non-faces (M=96%; p<.01). Reaction time differed between faces and non-faces (1.6s vs 2.2s; p=.014). The average local-global precedence score (LGP; local RT- global RT) was 41.4ms (SD=199.2ms) suggesting that, overall, there was no significant local or global precedence effect for the group (i.e. LGP was not significantly different from zero; t=1.8, d.f.=26, p=.29). No significant associations were found between measures of the Navon task and the perceptual closure task performance.
fMRI
The fMRI data showed increased activation for faces relative to non-faces in the right fusiform face area (FFA), bilateral inferior frontal gyri, bilateral parietal lobule, left amygdala and right angular gyrus and insula (FDR corrected: p<.05; >10 contiguous voxels). No age-related effects were found for behavioral and neuroimaging data.

Conclusions:
The behavioral data suggest that face recognition is more dependent on holistic processing than non-face recognition or global-local interference processing. This finding is further strengthened by the large difference in detection rates in the self-paced task. The fMRI data showed typical responses in the FFA and inferior frontal gyri consistent with other recent studies demonstrating increased activation for faces compared to non-faces in autism. Overall, these findings suggest a specific face recognition deficit in adults and elderly with ASD independent of age. While informative, future analyses will include a matched healthy control group to elucidate whether these results are specific for ASD.