Sleep and Attention in Children with ASD-Associated Gene-Disruptive De Novo Mutations and Idiopathic ASD

Background: Sleep difficulties are common in autism spectrum disorder (ASD) and are especially evident in children with ASD-associated gene-disruptive de novo mutations (LGDMs), such as CHD8 and DYRK1A. Despite extensive knowledge of comorbid sleep issues in ASD, little is known about the complex interplay between the genetic etiologies of ASD, sleep, and daytime attentional skills.

Objectives: We had two primary aims: (1) To characterize neural patterns of attention associated with parent-reported sleep issues; (2) To assess whether these patterns were more severe for children with an ASD-associated LGDM. We hypothesized that children with ASD and sleep problems may showcase poor attention, as evident by reduced discrimination between frequent and novel conditions and faster patterns of habituation on the Novelty P300 experiment. For children with ASD-associated LGDM we expected this pattern to be more robust.

Methods: Participants (N = 95) included children with an ASD-associated LGDM and idiopathic children with ASD without a known LGDM (NON; see Table 1 for characterization). Child attention was indexed using EEG while children watched a video and passively listening to sounds in an auditory oddball paradigm with a frequent condition (70%), an infrequent pitch deviant condition (15%), and an infrequent novelty deviant condition (15%). We focused on condition differences between the frequent and novelty deviant condition for the central P3a ERP component (180-350 ms). We used a single-trial approach to track dynamic patterns of attention (i.e., the variability of the P3a over the course of the experiment). Caregivers reported current sleep problems during a structured interview that included subdomains of daytime tiredness, problems with sleep onset, and nighttime awakenings.

Results: Multilevel models revealed a significant interaction between condition and sleep problems, F (1, 9156) = 8.03, p = .004, indicating less condition discrimination for children with sleep problems. For children with sleep problems, novelty habituation was more rapid (i.e., amplitude reducing more quickly) for children with sleep problems, F (1, 137) = 3.18, p = .07 (see Figure 1). There were no ERP differences between LGDM and NON among children without sleep problems. However, among those with sleep problems, LGDM children exhibited reduced condition discrimination relative to NON, especially in children who exhibit daytime tiredness and difficulties falling asleep.

Conclusions: These results confirm prior reports that sleep problems interfere with neural mechanisms of daytime attention and extend our understanding to dynamic patterns of daytime attention. Specifically, reduced novelty discrimination (i.e., attention orienting in response to infrequent, novel sounds) in children with ASD and sleep problems is in part due to rapid novelty habituation. Consequences of sleep problems were evident in both NON and LGDM groups; however, children with a LGDM did exhibit more severe attention disruptions particularly related to daytime tiredness and difficulties falling asleep. Further exploration of sleep problems in genetic subtypes of ASD, such as CHD8 and DYRK1A, can inform the interaction between sleep and etiological contributions to ASD. We will discuss the biological implications at the intersection of sleep problems, dynamic attention, and genetic subtypes of ASD.