Abstract:

Autism spectrum disorder (ASD) individuals are confirmed to have deficits in spatial working memory, but the specific impaired element is still undefined. Chunking is an effective form of strategic encoding that involves the recoding of a set of data into a compressed, efficient form and can extend working memory capacity. Spatial chunking is a key segment of the spatial working memory processing. Structured stimuli significantly encourage chunking, lessening working memory demand and significantly improving working memory performance. In this study, we used the Sternberg spatial working memory task and the Corsi’s spatial span task, manipulated the structure of stimulus’ spatial arrangement. In this method, we investigated the chunking processing of autism preschool children, and evaluated the relationship between spatial chunking and spatial working memory capacity in ASD. All participants were recruited in a integrated kindergarten, including 15 ASD children and 18 typical developing children. Age range from 4.5 to 7 years old. Participants were matched in age, verbal intelligence and nonverbal intelligence. In experiment1, participants were required to remember several stimulus’ positions which are presented simultaneously, and only judge one position of them. The accuracy of participants were recorded. In experiment2, stimulus participants were presented one by one, participants were required to remember several stimulus’ positions, and repeat the stimulus positions in the same order. Experiments were manipulated into 2 conditions. In the high-structured condition, all sequences followed a structured rule such that locations were organized as familiar shape components, involving symmetry and parallel sides, and were thus more easily organized into higher-level patterns. In the low-structured condition, locations were organized as irregular figures. The results in experiment1 proved that ASD children existing chunking deficits. First of all, healthy children performed significant better than ASD Children in high-structured condition [F(1,31) = 18.21, p < 0.001, ηp2 = 0.370]. But there was no difference in low-structured condition between the 2 groups [F(1,31) = 0.87, p = 0.358, ηp2 = 0.027]. Besides, healthy children performed significant better in high-structured condition comparing with the low-structured condition (p < 0.001). In contrast, ASD Children didn’t show any difference in the 2 conditions (p = 0.245). This proved that the high-structured arrangement encouraged high-level chunking in typical developing children. But ASD children were insensitive in the structured arrangement of stimulus and can’t apply chunking strategy effectively. According to the results in experiment2, the spatial capacity of ASD children was significantly lower than healthy children no matter in high- or low-conditions [F(1,31) = 41.68, p < 0.001, ηp2 = 0.573; F(1,31) = 12.65, p = 0.001, ηp2 = 0.290]. High-structured arrangement significantly improved the spatial working memory capacity in typical developing children (p < 0.001), but also influenced ASD children to some extent (p = 0.041). Chunking deficits influenced spatial capacity in ASD children, abnormal memory storage is also an important factor. In summary, ASD children showed impairment in chunking in spatial working memory, cognitive style of weak central coherence plays a crucial role in spatial chunking. Besides, attention control and executive function have indirect effect in spatial working memory processing in other domains. These findings not only provide inspiration for developing cognitive intervention model, but also provide a breakthrough to the neurophysiological mechanism studies in ASD.

Key words: autism spectrum disorder, spatial working memory, chunking, weak central coherence