Selective Attention and Context Memory in Adults with ASD

Background:

Previous evidence suggests intact item memory but impaired context memory in adults with ASD. Similar findings have been observed in healthy older adults (HOA) and thought to be related to deficits in relational binding (i.e., forming an association between the item and context). However, this deficit can be overcome when attention is directed towards single item-context associations (James et al., 2016; Powell et al., 2018). While many studies used aging analogies to describe the pattern of memory performance in ASD, few have directly compared performance to HOA (Bowler, et al., 2004; 2007; Ring et al., 2016).

Objectives:

To address whether context memory performance in adults with ASD can be characterized as ‘cognitively old’, and whether it can be improved when attention is directed towards single item-context associations.

Methods:

Participants were 22 young adults (ages 18–35), 18 HOA (ages 60–80), and 26 adults with ASD, ages 20–58. Participants completed an experimental task designed to assess item memory and context memory. During the study phase, participants were shown a centrally-presented object in the presence of two contextual features: a color and a scene positioned to right and left of object. For each trial, participants directed their attention to the object's relationship with one of the two contextual features (Target Context) while ignoring the other (Distractor Context). EEG was recorded throughout the experiment.

Results:

Memory for the centrally-presented object (item memory) indicated similar performance across diagnostic groups (Younger: M = .73; Older: M = .71; ASD: M = .62), F(1,49) = 1.71, p = .20, η_p² = .07. Next, the proportion of correctly recognized Target and Distractor Contexts was examined using a 2 (Target vs. Distractor) by 3 (Diagnostic Group: ASD vs. Younger vs. Older) Repeated Measures ANOVA. This analysis demonstrated a main effect of context indicating greater memory for the target context relative to the distractor context (M = .66, SD = .11; M = .52, SD = .04, respectively), F(1,63) = 92.66, p < .001, η_p² = .65. However this main effect was qualified by a significant context by diagnostic group interaction, F(2,63) = 8.56, p = .001, η_p² = .26. Post-hoc comparisons indicated target context memory was significantly greater in younger adults (M = .74, SD = .08) relative to HOA (M = .63, SD = .06) and adults with ASD (M = .59, SD = .12). There was no difference in performance between adults with ASD and HOA.

Conclusions:

The current findings suggest that context memory in adults with ASD can be improved when attention is directed to single item-context associations. However, the level of context memory performance in adults with ASD was more comparable to the level of performance seen in HOA than younger adults, suggesting a ‘cognitively old’ pattern of context memory performance. Given that the majority of current sample of adults with ASD were less than 45 years old, these findings may have important implications for the specific cognitive abilities that may be more disrupted in adults with ASD as they age.