ASD-Related Memory Dysfunction in Middle Childhood

Background: Although individuals with autism spectrum disorder (ASD) display similar memory performance on semantic memory tasks compared to typically developing controls (TDC), they perform significantly more poorly when recalling episodic events, which might derive from altered hippocampal function. If this is the case, we should observe memory deficits when children are asked to retrieve information about the association between an event and its context, even with tasks that reduce the strategic memory process demands.

Objectives: To compare memory for event-context association in children with ASD and TDC. Item-color context versus an item-space context associations were examined.

Methods: Participants included 62 children with ASD and 68 TDC (Table 1). The ability to recall events in association with specific contextual details was assessed across two tasks. In the event-color task, 80 black-ink drawings were presented with a border in one of 4 colors (20 of each color in a random order). Participants were instructed to try to remember both the item and the color of the border. The spatial position task was identical, except that the drawings did not vary in colored border but in the spatial position in which they appeared (1 of the 4 quadrants of the computer screen in random order). Rate of accurate recollection of item-context association, and discrimination between old and new items irrespective of context (i.e., d’ prime) were calculated.

Results: We conducted a 2 (Diagnosis: ASD versus TDC) X 2 (Detail type: color versus spatial) mixed ANOVA with rate of correct event-context associations as the dependent measure. We found a significant main effect of diagnosis F(1, 115) = 16.81, p < .001, n² = .06, indicating a significant difference between ASD (M = 0.52, SD =0.22) and TDC (M = 0.65, SD =0.23) across both conditions. There was also a significant main effect of detail type, F(1, 115) = 8.29, p < .01, n² = .03, such that children recollected fewer item-color (M = 0.55, SD =0.23) compared to item-space associations (M = 0.63, SD =0.23). Critically, there was also a significant diagnosis by detail interaction, F(1, 119) = 9.22, p < .01, n² = .03, (Figure 1), suggesting that the ASD group exhibited worse performance for the event-space condition. This interaction persisted when we included d’ and IQ as covariates, F(1, 116) = 12.00, p < .001, n² = .03.

Conclusions: Event-context deficits were restricted to the spatial aspect of episodic memory even after controlling for IQ and the capacity to discriminate new images from older images. Discussion will center on possible explanations for this specific deficit, which might include reduced reliance on conceptual representations of space.