The nature of integrative processing problems in individuals with autism spectrum disorder

Abstract

Background: The Weak Central Coherence (WCC) account of autism spectrum disorder (ASD) proposes a weakness of global integrative processing (i.e. central coherence) resulting in a strength in local, detailed, focused processing in individuals with ASD. However, because of the mixed findings regarding weaknesses in global integrative processing, recent literature had neglected this part of the WCC theory. Mixed findings about global integrative deficits in ASD came about because WCC was operationalized differently in different studies. This research’s primary aim was to delineate the nature of integrative processing that is impaired in ASD more precisely. Integrative processes were demarcated into first order and second order processes. Individuals with ASD were hypothesized to have problems in second order integrative processing only, which refers to the apprehension of inter-elemental relations that exist external to the individual elements. The hypothesis was investigated in both the visual-perceptual domain (Study One) and the verbal-conceptual domain (Study Two). In the visual-perceptual domain, first order and second order integrative processing refer to global processing and gestalt processing respectively. In the verbal-conceptual domain, they refer to the apprehension of taxonomic relations and thematic relations respectively. With better delineation of the construct of integrative processing in WCC, Study Three pursued the secondary aims of this research: to use the clarified constructs to study whether WCC exists as a central mechanism, and to test its predictive value on ASD symptomatology.

Methods: Twenty high functioning adolescents with ASD (HFA, aged 12 to 15) and 20 matched typically developing (TD) counterparts (aged 12 to 15) participated in the three studies of this research. In Study One, participants were administered a first order task with hierarchical compound stimuli that aimed to tap their global processing, and a second order task that aimed to tap their apprehension of gestalt principles (gestalt processing). In Study Two, participants were administered a lexical decision priming task with half of the prime-target pairs denoting a taxonomic relation (first order task), and half of the prime-target pairs denoting a thematic relation (second order task). In Study Three, correlation analyses were done among the second order task measures of the two domains as well as HFA symptomatology indexed by the Autism Quotient (AQ).

Results: For Study One, there was a trend showing that participants with HFA performed worse than TD peers in a subtask that tapped one gestalt principle (the principle of similarity). Contrary to prediction, in the global processing task, HFA individuals exhibited a reliable local bias. For Study Two, HFA participants were found to be primed to a significantly lesser extent by a thematic prime while exhibiting intact taxonomic priming. In Study Three, cross domain associations of second order processing measures were not significant, which was against the notion of a central mechanism of WCC. The priming extent by a thematic prime and a measure of gestalt processing were found to associate negatively with ASD symptoms in the TD group.

Conclusions: All in all, the present research had partial success in clarifying the nature of WCC as a weakness in second order integrative processing. Theoretical and practical significance as well as future research directions were discussed.