Perturbed cortical hierarchies in Autism Spectrum Disorder: The case of high-level vision.

While people with Autism Spectrum Disorder (ASD) show clear impairments in normal day-to-day behavior, they simultaneously outperform their Typically Developing (TD) counterparts on many perceptual tasks. This relative superiority in tasks requiring the processing of separate features rather than the stimulus configuration, evoked two competing theoretical frameworks: the Weak Central Coherence hypothesis (WCC) and the Enhanced Perceptional Functioning (EPF) theory. While the former explains these results by the absence of a normal bias towards the global structure in a visual display in people with ASD, the latter insists that generally increased attention for local elements in a scene is the cause. Although both theoretical frameworks (WCC and EPF) have been investigated for years, evidence for either line of theory remains mixed and ambiguously interpretable. To address this problem, we argue that the existing evidence on perceptual processing in ASD should be integrated within a unified theoretical framework, the Reverse Hierarchy Theory (RHT). Within this theory, both neurocognitive theories can be regarded as cases of hierarchy perturbations in the visual cortex. More precisely, people with ASD show a less efficient (slower) early-stage and pre-attentive global processing of visual information (in accordance with WCC) and a later-stage advantage in the attention-focused processing of local scene (or object) elements (in accordance with EPF).

            In this PhD project, we evaluated previous research findings on coarse (global) and fine (local) visual processing in ASD and tested the critical ideas of the main neurocognitive theories (WCC and EPF) in innovative ways. We used a diverse set of time-dependent experimental paradigms to explore possible differences in high-level visual perception between adolescents and adults with and without ASD (12-25 years of age). More precisely, behavioral testing focused on specific combinations of five experimental topics of interest: (1) the use of more ecologically valid visual displays, (2) the role of attentional processes in perception, (3) the defining stimulus characteristics for coarse and fine visual processing, (4) the influence of socially relevant stimuli, and (5) the influence of implicit social inferences based on the perception of dynamic visual displays.

            We believe that, by linking current neurocognitive models with state-of-the-art research on the processing sequence of coarse and fine information in TD participants, RHT provides an important addition to our current understanding of visual processing in ASD. Future research should continue to focus on systematically varying the time-course and content of stimulus presentation in behavioral paradigms and to link it with neuroimaging findings. As a result, we believe that the ASD-related literature will converge on a neurologically plausible and coherent theoretical framework for understanding the atypical visual perception in ASD.